{
  "nbformat": 4,
  "nbformat_minor": 0,
  "metadata": {
    "colab": {
      "provenance": []
    },
    "kernelspec": {
      "name": "python3",
      "display_name": "Python 3"
    },
    "language_info": {
      "name": "python"
    }
  },
  "cells": [
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "id": "Nhpz55dkcEc2"
      },
      "outputs": [],
      "source": [
        "# Import library yang diperlukan\n",
        "import numpy as np\n",
        "import matplotlib.pyplot as plt\n",
        "from keras.datasets import mnist\n",
        "from keras.models import Sequential\n",
        "from keras.layers import Conv2D, MaxPooling2D, Flatten, Dense\n",
        "from keras.utils import to_categorical"
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "# Memuat dataset MNIST sebagai contoh\n",
        "(X_train, y_train), (X_test, y_test) = mnist.load_data()\n",
        "\n",
        "# Visualisasi data sebelum proses CNN\n",
        "plt.figure(figsize=(12, 4))\n",
        "for i in range(5):\n",
        "    plt.subplot(1, 5, i+1)\n",
        "    plt.imshow(X_train[i], cmap='gray')\n",
        "    plt.axis('off')\n",
        "plt.suptitle('Data Latih Sebelum CNN')\n",
        "\n",
        "# Normalisasi dan reshape data\n",
        "X_train = X_train.reshape(-1, 28, 28, 1) / 255.0\n",
        "X_test = X_test.reshape(-1, 28, 28, 1) / 255.0\n",
        "y_train = to_categorical(y_train, 10)\n",
        "y_test = to_categorical(y_test, 10)\n",
        "\n",
        "# Membangun model CNN\n",
        "model = Sequential([\n",
        "    Conv2D(32, (3, 3), activation='relu', input_shape=(28, 28, 1)),\n",
        "    MaxPooling2D((2, 2)),\n",
        "    Flatten(),\n",
        "    Dense(128, activation='relu'),\n",
        "    Dense(10, activation='softmax')\n",
        "])\n",
        "\n",
        "# Kompilasi model\n",
        "model.compile(optimizer='adam', loss='categorical_crossentropy', metrics=['accuracy'])\n",
        "\n",
        "# Melatih model\n",
        "history = model.fit(X_train, y_train, epochs=5, batch_size=64, validation_split=0.2)\n",
        "\n",
        "plt.tight_layout()\n",
        "\n",
        "# Evaluasi model pada data uji\n",
        "test_loss, test_acc = model.evaluate(X_test, y_test)\n",
        "print(f'Test accuracy: {test_acc:.4f}')\n",
        "\n",
        "# Prediksi dan visualisasi model\n",
        "predictions = model.predict(X_test)\n",
        "\n",
        "# Visualisasi data setelah proses CNN\n",
        "num_samples = 5\n",
        "plt.figure(figsize=(12, 4))\n",
        "for i in range(num_samples):\n",
        "    plt.subplot(1, num_samples, i+1)\n",
        "    plt.imshow(X_test[i].reshape(28, 28), cmap='gray')\n",
        "    true_label = np.argmax(y_test[i])\n",
        "    predicted_label = np.argmax(predictions[i])\n",
        "    plt.title(f\"True: {true_label}, Predicted: {predicted_label}\")\n",
        "    plt.axis('off')\n",
        "plt.suptitle('Data Uji Setelah CNN')\n",
        "\n",
        "plt.tight_layout()\n",
        "plt.show()"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 955
        },
        "id": "rBmkD84tcKR6",
        "outputId": "4cb5a42e-acf1-4488-d906-d40952884ace"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Downloading data from https://storage.googleapis.com/tensorflow/tf-keras-datasets/mnist.npz\n",
            "11490434/11490434 [==============================] - 0s 0us/step\n",
            "Epoch 1/5\n",
            "750/750 [==============================] - 40s 51ms/step - loss: 0.1969 - accuracy: 0.9419 - val_loss: 0.0774 - val_accuracy: 0.9773\n",
            "Epoch 2/5\n",
            "750/750 [==============================] - 38s 51ms/step - loss: 0.0649 - accuracy: 0.9811 - val_loss: 0.0605 - val_accuracy: 0.9835\n",
            "Epoch 3/5\n",
            "750/750 [==============================] - 30s 40ms/step - loss: 0.0422 - accuracy: 0.9870 - val_loss: 0.0570 - val_accuracy: 0.9836\n",
            "Epoch 4/5\n",
            "750/750 [==============================] - 29s 38ms/step - loss: 0.0291 - accuracy: 0.9910 - val_loss: 0.0518 - val_accuracy: 0.9843\n",
            "Epoch 5/5\n",
            "750/750 [==============================] - 30s 39ms/step - loss: 0.0208 - accuracy: 0.9937 - val_loss: 0.0562 - val_accuracy: 0.9840\n",
            "313/313 [==============================] - 2s 6ms/step - loss: 0.0466 - accuracy: 0.9841\n",
            "Test accuracy: 0.9841\n",
            "313/313 [==============================] - 2s 6ms/step\n"
          ]
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 1200x400 with 5 Axes>"
            ],
            "image/png": "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\n"
          },
          "metadata": {}
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 1200x400 with 5 Axes>"
            ],
            "image/png": "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\n"
          },
          "metadata": {}
        }
      ]
    }
  ]
}