autoencoder_mnist_handwritten_digit.ipynb

{
  "cells": [
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "id": "1YfqRHaUIQZN"
      },
      "outputs": [],
      "source": [
        "import numpy as np\n",
        "import matplotlib.pyplot as plt\n",
        "from keras.layers import Input, Dense\n",
        "from keras.models import Model\n",
        "from keras.datasets import mnist\n",
        "from keras.utils import to_categorical\n",
        "from keras.optimizers import Adam\n",
        "from sklearn.metrics import confusion_matrix\n",
        "import seaborn as sns"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {},
      "source": [
        "##### **Dataset**"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "-Xlla3ndPzNl",
        "outputId": "37a93dc5-b3e1-45c4-bd94-ec0caacc29f5"
      },
      "outputs": [
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "Downloading data from https://storage.googleapis.com/tensorflow/tf-keras-datasets/mnist.npz\n",
            "11490434/11490434 [==============================] - 0s 0us/step\n",
            "Training data dimensions: (60000, 28, 28)\n",
            "Testing data dimensions: (10000, 28, 28)\n"
          ]
        }
      ],
      "source": [
        "# Load the MNIST dataset\n",
        "(train_images, train_labels), (test_images, test_labels) = mnist.load_data()\n",
        "\n",
        "# Print dimensions\n",
        "print(f\"Training data dimensions: {train_images.shape}\")\n",
        "print(f\"Testing data dimensions: {test_images.shape}\")"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {},
      "source": [
        "It means that 60k 28x28 pictures exist in training data and 10k 28x28 pictures exist in testing (validation) data."
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {},
      "source": [
        "- display sample images:"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 1000
        },
        "id": "Lef-K2i0QQ0H",
        "outputId": "e320a89c-49fb-483e-d27c-37f97bd3041a"
      },
      "outputs": [
        {
          "data": {
            "image/png": "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",
            "text/plain": [
              "<Figure size 800x200 with 5 Axes>"
            ]
          },
          "metadata": {},
          "output_type": "display_data"
        },
        {
          "data": {
            "image/png": "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",
            "text/plain": [
              "<Figure size 800x200 with 5 Axes>"
            ]
          },
          "metadata": {},
          "output_type": "display_data"
        },
        {
          "data": {
            "image/png": "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",
            "text/plain": [
              "<Figure size 800x200 with 5 Axes>"
            ]
          },
          "metadata": {},
          "output_type": "display_data"
        },
        {
          "data": {
            "image/png": "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",
            "text/plain": [
              "<Figure size 800x200 with 5 Axes>"
            ]
          },
          "metadata": {},
          "output_type": "display_data"
        },
        {
          "data": {
            "image/png": "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",
            "text/plain": [
              "<Figure size 800x200 with 5 Axes>"
            ]
          },
          "metadata": {},
          "output_type": "display_data"
        },
        {
          "data": {
            "image/png": "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",
            "text/plain": [
              "<Figure size 800x200 with 5 Axes>"
            ]
          },
          "metadata": {},
          "output_type": "display_data"
        },
        {
          "data": {
            "image/png": "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",
            "text/plain": [
              "<Figure size 800x200 with 5 Axes>"
            ]
          },
          "metadata": {},
          "output_type": "display_data"
        },
        {
          "data": {
            "image/png": "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",
            "text/plain": [
              "<Figure size 800x200 with 5 Axes>"
            ]
          },
          "metadata": {},
          "output_type": "display_data"
        },
        {
          "data": {
            "image/png": "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",
            "text/plain": [
              "<Figure size 800x200 with 5 Axes>"
            ]
          },
          "metadata": {},
          "output_type": "display_data"
        },
        {
          "data": {
            "image/png": "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",
            "text/plain": [
              "<Figure size 800x200 with 5 Axes>"
            ]
          },
          "metadata": {},
          "output_type": "display_data"
        }
      ],
      "source": [
        "# Display five images from each class\n",
        "for digit in range(10):\n",
        "    digit_indices = np.where(train_labels == digit)[0]\n",
        "    sample_indices = digit_indices[:5]  # Get the first five samples\n",
        "    plt.figure(figsize=(8, 2))\n",
        "    for i, sample_index in enumerate(sample_indices):\n",
        "        plt.subplot(1, 5, i + 1)\n",
        "        plt.imshow(train_images[sample_index], cmap=\"gray\")\n",
        "        plt.title(f\"Digit {digit}\")\n",
        "        plt.axis(\"off\")\n",
        "\n",
        "plt.show()\n"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {},
      "source": [
        "---"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {},
      "source": [
        "- MNIST Digit Class Distribution for training data histogram:"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 472
        },
        "id": "P8uMcawecCHA",
        "outputId": "368e4a21-5c7f-4ad7-d58c-568c5a156e8e"
      },
      "outputs": [
        {
          "data": {
            "image/png": "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",
            "text/plain": [
              "<Figure size 640x480 with 1 Axes>"
            ]
          },
          "metadata": {},
          "output_type": "display_data"
        }
      ],
      "source": [
        "# Count occurrences of each class label for training data\n",
        "class_counts = np.bincount(train_labels)\n",
        "\n",
        "# Create a bar plot\n",
        "plt.bar(range(10), class_counts)\n",
        "plt.xlabel(\"Digit Class\")\n",
        "plt.ylabel(\"Number of Samples\")\n",
        "plt.title(\"MNIST Digit Class Distribution for training data\")\n",
        "plt.xticks(range(10), labels=[str(i) for i in range(10)])\n",
        "plt.show()\n"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {},
      "source": [
        "- MNIST Digit Class Distribution for testing data histogram:"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 472
        },
        "id": "CP2_lTROm6k2",
        "outputId": "c971b963-7c60-4ec2-aa87-11d6b084b44f"
      },
      "outputs": [
        {
          "data": {
            "image/png": "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",
            "text/plain": [
              "<Figure size 640x480 with 1 Axes>"
            ]
          },
          "metadata": {},
          "output_type": "display_data"
        }
      ],
      "source": [
        "# Count occurrences of each class label for testing data\n",
        "class_counts = np.bincount(test_labels)\n",
        "\n",
        "# Create a bar plot\n",
        "plt.bar(range(10), class_counts)\n",
        "plt.xlabel(\"Digit Class\")\n",
        "plt.ylabel(\"Number of Samples\")\n",
        "plt.title(\"MNIST Digit Class Distribution for testing data\")\n",
        "plt.xticks(range(10), labels=[str(i) for i in range(10)])\n",
        "plt.show()"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {},
      "source": [
        "we don't need any process for data balancing, because the number of samples for classes is close to each other."
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {},
      "source": [
        "---"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {},
      "source": [
        "- Normalize data using `min-max normalization` to scale [0,1]\n",
        "\n",
        "  - why we use Min-Max normalization?\n",
        "\n",
        "  1. Enhances Model Performance: Normalized data enhances model performance and improves the accuracy of a model.\n",
        "\n",
        "  2. Stability in Optimization Process: Normalization fosters stability in the optimization process, promoting faster convergence during gradient-based training.\n",
        "\n",
        "  3. Easy Interpretation: Normalized data is also easy to interpret and thus, easier to understand. When all the features of a dataset are on the same scale, it also becomes easier to identify.\n"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 472
        },
        "id": "ypeSH19ge8Bp",
        "outputId": "430442bf-1b29-4e82-f23a-fda0a390c49d"
      },
      "outputs": [
        {
          "data": {
            "image/png": "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",
            "text/plain": [
              "<Figure size 640x480 with 1 Axes>"
            ]
          },
          "metadata": {},
          "output_type": "display_data"
        }
      ],
      "source": [
        "# Flatten the 28x28 images into vectors of size 784\n",
        "train_X_flat = train_images.reshape(-1, 784)\n",
        "test_X_flat = test_images.reshape(-1, 784)\n",
        "\n",
        "# ploting Pixel Intensity before normalization\n",
        "plt.hist(train_images[0], bins=20)\n",
        "plt.xlabel(\"Pixel Intensity\")\n",
        "plt.ylabel(\"Frequency\")\n",
        "plt.title(\"Histogram of MNIST Pixel Intensities\")\n",
        "plt.show()\n"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 472
        },
        "id": "Ic8DZpYw7UH7",
        "outputId": "54e051c9-8077-400a-d432-aba5a678fb18"
      },
      "outputs": [
        {
          "data": {
            "image/png": "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",
            "text/plain": [
              "<Figure size 640x480 with 1 Axes>"
            ]
          },
          "metadata": {},
          "output_type": "display_data"
        }
      ],
      "source": [
        "# Normalize data using min-max normalization to scale [0,1]\n",
        "normalized_train_X_flat = train_X_flat.astype('float32') / 255.\n",
        "normalized_test_X_flat = test_X_flat.astype('float32') / 255.\n",
        "\n",
        "# ploting Pixel Intensity after normalization\n",
        "plt.hist(normalized_train_X_flat[0], bins=20, range=(0, 1))\n",
        "plt.xlabel(\"Normalized Pixel Intensity\")\n",
        "plt.ylabel(\"Frequency\")\n",
        "plt.title(\"Histogram of Normalized MNIST Pixel Intensities\")\n",
        "plt.show()"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {},
      "source": [
        "---\n"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {},
      "source": [
        "##### **Auto-Encoder Network**\n",
        "\n",
        "  Input and output size of this net is 784 because $784 = 28^{2}$ (we use it for reconstructing input).\n",
        "\n",
        "  We use `sigmoid` activation function for the last layer of decoder part to reconstructing input and `relu` for other layers.\n",
        "\n",
        "  The loss function is `binary crossentropy` because there is 2 state for output: reconstructing in correct way or incorrect.\n",
        "\n",
        "  We use `Adam` Optimizer because it's more common!"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "id": "6IHvXH8r8dr9"
      },
      "outputs": [],
      "source": [
        "# Define the encoder layers\n",
        "input_img = Input(shape=(784,))\n",
        "encoded = Dense(500, activation='relu')(input_img)\n",
        "encoded = Dense(100, activation='relu')(encoded)\n",
        "encoded = Dense(30, activation='relu')(encoded)\n",
        "\n",
        "# Define the decoder layers\n",
        "decoded = Dense(100, activation='relu')(encoded)\n",
        "decoded = Dense(500, activation='relu')(decoded)\n",
        "decoded = Dense(784, activation='sigmoid')(decoded)"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "id": "ZHnwS4M88weL"
      },
      "outputs": [],
      "source": [
        "# Define the autoencoder model\n",
        "autoencoder = Model(input_img, decoded)\n",
        "\n",
        "# Compile the model\n",
        "autoencoder.compile(optimizer=Adam(), loss='binary_crossentropy')"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "collapsed": true,
        "id": "dtzEUSeK4URa",
        "outputId": "e54c2f65-a8c1-48a0-91e0-182fc4fe9bca"
      },
      "outputs": [
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "Epoch 1/50\n",
            "235/235 [==============================] - 11s 41ms/step - loss: 0.2141 - val_loss: 0.1419\n",
            "Epoch 2/50\n",
            "235/235 [==============================] - 11s 48ms/step - loss: 0.1264 - val_loss: 0.1140\n",
            "Epoch 3/50\n",
            "235/235 [==============================] - 11s 45ms/step - loss: 0.1102 - val_loss: 0.1048\n",
            "Epoch 4/50\n",
            "235/235 [==============================] - 10s 42ms/step - loss: 0.1032 - val_loss: 0.0997\n",
            "Epoch 5/50\n",
            "235/235 [==============================] - 9s 39ms/step - loss: 0.0988 - val_loss: 0.0959\n",
            "Epoch 6/50\n",
            "235/235 [==============================] - 11s 46ms/step - loss: 0.0958 - val_loss: 0.0938\n",
            "Epoch 7/50\n",
            "235/235 [==============================] - 11s 45ms/step - loss: 0.0936 - val_loss: 0.0925\n",
            "Epoch 8/50\n",
            "235/235 [==============================] - 11s 46ms/step - loss: 0.0920 - val_loss: 0.0907\n",
            "Epoch 9/50\n",
            "235/235 [==============================] - 8s 35ms/step - loss: 0.0907 - val_loss: 0.0903\n",
            "Epoch 10/50\n",
            "235/235 [==============================] - 11s 45ms/step - loss: 0.0897 - val_loss: 0.0889\n",
            "Epoch 11/50\n",
            "235/235 [==============================] - 10s 44ms/step - loss: 0.0887 - val_loss: 0.0881\n",
            "Epoch 12/50\n",
            "235/235 [==============================] - 10s 42ms/step - loss: 0.0880 - val_loss: 0.0875\n",
            "Epoch 13/50\n",
            "235/235 [==============================] - 9s 39ms/step - loss: 0.0874 - val_loss: 0.0869\n",
            "Epoch 14/50\n",
            "235/235 [==============================] - 10s 43ms/step - loss: 0.0868 - val_loss: 0.0865\n",
            "Epoch 15/50\n",
            "235/235 [==============================] - 10s 44ms/step - loss: 0.0863 - val_loss: 0.0859\n",
            "Epoch 16/50\n",
            "235/235 [==============================] - 9s 36ms/step - loss: 0.0859 - val_loss: 0.0857\n",
            "Epoch 17/50\n",
            "235/235 [==============================] - 10s 44ms/step - loss: 0.0853 - val_loss: 0.0854\n",
            "Epoch 18/50\n",
            "235/235 [==============================] - 10s 44ms/step - loss: 0.0850 - val_loss: 0.0851\n",
            "Epoch 19/50\n",
            "235/235 [==============================] - 11s 46ms/step - loss: 0.0847 - val_loss: 0.0851\n",
            "Epoch 20/50\n",
            "235/235 [==============================] - 9s 38ms/step - loss: 0.0843 - val_loss: 0.0847\n",
            "Epoch 21/50\n",
            "235/235 [==============================] - 10s 44ms/step - loss: 0.0840 - val_loss: 0.0841\n",
            "Epoch 22/50\n",
            "235/235 [==============================] - 10s 42ms/step - loss: 0.0837 - val_loss: 0.0841\n",
            "Epoch 23/50\n",
            "235/235 [==============================] - 10s 45ms/step - loss: 0.0834 - val_loss: 0.0835\n",
            "Epoch 24/50\n",
            "235/235 [==============================] - 9s 40ms/step - loss: 0.0831 - val_loss: 0.0833\n",
            "Epoch 25/50\n",
            "235/235 [==============================] - 10s 44ms/step - loss: 0.0827 - val_loss: 0.0830\n",
            "Epoch 26/50\n",
            "235/235 [==============================] - 10s 43ms/step - loss: 0.0825 - val_loss: 0.0829\n",
            "Epoch 27/50\n",
            "235/235 [==============================] - 8s 36ms/step - loss: 0.0823 - val_loss: 0.0828\n",
            "Epoch 28/50\n",
            "235/235 [==============================] - 10s 43ms/step - loss: 0.0820 - val_loss: 0.0826\n",
            "Epoch 29/50\n",
            "235/235 [==============================] - 10s 44ms/step - loss: 0.0819 - val_loss: 0.0826\n",
            "Epoch 30/50\n",
            "235/235 [==============================] - 9s 39ms/step - loss: 0.0817 - val_loss: 0.0824\n",
            "Epoch 31/50\n",
            "235/235 [==============================] - 10s 41ms/step - loss: 0.0815 - val_loss: 0.0824\n",
            "Epoch 32/50\n",
            "235/235 [==============================] - 10s 43ms/step - loss: 0.0814 - val_loss: 0.0822\n",
            "Epoch 33/50\n",
            "235/235 [==============================] - 10s 42ms/step - loss: 0.0812 - val_loss: 0.0820\n",
            "Epoch 34/50\n",
            "235/235 [==============================] - 9s 37ms/step - loss: 0.0811 - val_loss: 0.0819\n",
            "Epoch 35/50\n",
            "235/235 [==============================] - 12s 51ms/step - loss: 0.0809 - val_loss: 0.0816\n",
            "Epoch 36/50\n",
            "235/235 [==============================] - 11s 46ms/step - loss: 0.0808 - val_loss: 0.0817\n",
            "Epoch 37/50\n",
            "235/235 [==============================] - 10s 44ms/step - loss: 0.0807 - val_loss: 0.0818\n",
            "Epoch 38/50\n",
            "235/235 [==============================] - 9s 39ms/step - loss: 0.0806 - val_loss: 0.0816\n",
            "Epoch 39/50\n",
            "235/235 [==============================] - 11s 46ms/step - loss: 0.0805 - val_loss: 0.0815\n",
            "Epoch 40/50\n",
            "235/235 [==============================] - 14s 59ms/step - loss: 0.0804 - val_loss: 0.0815\n",
            "Epoch 41/50\n",
            "235/235 [==============================] - 23s 97ms/step - loss: 0.0803 - val_loss: 0.0817\n",
            "Epoch 42/50\n",
            "235/235 [==============================] - 17s 74ms/step - loss: 0.0802 - val_loss: 0.0813\n",
            "Epoch 43/50\n",
            "235/235 [==============================] - 14s 59ms/step - loss: 0.0801 - val_loss: 0.0813\n",
            "Epoch 44/50\n",
            "235/235 [==============================] - 10s 44ms/step - loss: 0.0800 - val_loss: 0.0810\n",
            "Epoch 45/50\n",
            "235/235 [==============================] - 13s 56ms/step - loss: 0.0800 - val_loss: 0.0810\n",
            "Epoch 46/50\n",
            "235/235 [==============================] - 12s 52ms/step - loss: 0.0799 - val_loss: 0.0811\n",
            "Epoch 47/50\n",
            "235/235 [==============================] - 13s 57ms/step - loss: 0.0798 - val_loss: 0.0813\n",
            "Epoch 48/50\n",
            "235/235 [==============================] - 11s 47ms/step - loss: 0.0798 - val_loss: 0.0810\n",
            "Epoch 49/50\n",
            "235/235 [==============================] - 13s 57ms/step - loss: 0.0796 - val_loss: 0.0807\n",
            "Epoch 50/50\n",
            "235/235 [==============================] - 16s 68ms/step - loss: 0.0796 - val_loss: 0.0810\n"
          ]
        }
      ],
      "source": [
        "# Train the model\n",
        "history = autoencoder.fit(normalized_train_X_flat, normalized_train_X_flat,\n",
        "                          epochs=50,\n",
        "                          batch_size=256,\n",
        "                          shuffle=True,\n",
        "                          validation_data=(normalized_test_X_flat, normalized_test_X_flat))"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {},
      "source": [
        "---"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {},
      "source": [
        "- Ploting the loss and validation loss"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 367
        },
        "id": "4BFCuzQm80d9",
        "outputId": "2fc85dc5-b954-4cc4-e9a2-2a5f4688205f"
      },
      "outputs": [
        {
          "data": {
            "image/png": "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",
            "text/plain": [
              "<Figure size 1000x500 with 1 Axes>"
            ]
          },
          "metadata": {},
          "output_type": "display_data"
        }
      ],
      "source": [
        "# Plot the loss and validation loss\n",
        "plt.figure(figsize=(10, 5))\n",
        "plt.plot(history.history['loss'], label='Loss')\n",
        "plt.plot(history.history['val_loss'], label='Validation Loss')\n",
        "plt.title('Model loss')\n",
        "plt.ylabel('Loss')\n",
        "plt.xlabel('Epoch')\n",
        "plt.legend()\n",
        "plt.show()"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {},
      "source": [
        "As the plot shows increasing epochs didn't have tangible effect after 45 epoch."
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {},
      "source": [
        "---"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {},
      "source": [
        "##### **classifier with encoder and 2 fully connected hidden layer**\n",
        "\n",
        "we use `softmax` activation function for output layer to obtain probability of input data being included in each class."
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "id": "dyUY6V3pZfhg"
      },
      "outputs": [],
      "source": [
        "# Create the encoder model\n",
        "encoder = Model(input_img, encoded)"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "bjJGkGjlaxZS",
        "outputId": "e3e1b7ad-f68c-4476-b277-61be35207da6"
      },
      "outputs": [
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "1875/1875 [==============================] - 7s 4ms/step\n",
            "313/313 [==============================] - 1s 3ms/step\n"
          ]
        }
      ],
      "source": [
        "# Use the encoder to extract the reduced features\n",
        "x_train_encoded = encoder.predict(normalized_train_X_flat)\n",
        "x_test_encoded = encoder.predict(normalized_test_X_flat)"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "id": "Hgv8zVrLZfVz"
      },
      "outputs": [],
      "source": [
        "# Define the classifier model\n",
        "input_encoded = Input(shape=(30,))\n",
        "hidden = Dense(64, activation='relu')(input_encoded)\n",
        "hidden = Dense(64, activation='relu')(hidden)\n",
        "output = Dense(10, activation='softmax')(hidden)\n",
        "classifier = Model(input_encoded, output)"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "QKh3p92UZfAN",
        "outputId": "57d7cec1-a0fb-469e-da7d-7da5994b869c"
      },
      "outputs": [
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "Epoch 1/50\n",
            "235/235 [==============================] - 3s 6ms/step - loss: 0.0755 - accuracy: 0.9760 - val_loss: 0.0867 - val_accuracy: 0.9727\n",
            "Epoch 2/50\n",
            "235/235 [==============================] - 1s 3ms/step - loss: 0.0729 - accuracy: 0.9772 - val_loss: 0.0870 - val_accuracy: 0.9725\n",
            "Epoch 3/50\n",
            "235/235 [==============================] - 1s 4ms/step - loss: 0.0713 - accuracy: 0.9780 - val_loss: 0.0949 - val_accuracy: 0.9702\n",
            "Epoch 4/50\n",
            "235/235 [==============================] - 1s 3ms/step - loss: 0.0706 - accuracy: 0.9777 - val_loss: 0.0919 - val_accuracy: 0.9713\n",
            "Epoch 5/50\n",
            "235/235 [==============================] - 1s 3ms/step - loss: 0.0678 - accuracy: 0.9788 - val_loss: 0.0842 - val_accuracy: 0.9730\n",
            "Epoch 6/50\n",
            "235/235 [==============================] - 1s 4ms/step - loss: 0.0679 - accuracy: 0.9787 - val_loss: 0.0794 - val_accuracy: 0.9745\n",
            "Epoch 7/50\n",
            "235/235 [==============================] - 1s 4ms/step - loss: 0.0642 - accuracy: 0.9795 - val_loss: 0.0803 - val_accuracy: 0.9750\n",
            "Epoch 8/50\n",
            "235/235 [==============================] - 2s 6ms/step - loss: 0.0659 - accuracy: 0.9792 - val_loss: 0.0867 - val_accuracy: 0.9742\n",
            "Epoch 9/50\n",
            "235/235 [==============================] - 1s 4ms/step - loss: 0.0623 - accuracy: 0.9804 - val_loss: 0.0815 - val_accuracy: 0.9733\n",
            "Epoch 10/50\n",
            "235/235 [==============================] - 1s 3ms/step - loss: 0.0613 - accuracy: 0.9809 - val_loss: 0.0799 - val_accuracy: 0.9741\n",
            "Epoch 11/50\n",
            "235/235 [==============================] - 1s 4ms/step - loss: 0.0615 - accuracy: 0.9805 - val_loss: 0.1010 - val_accuracy: 0.9671\n",
            "Epoch 12/50\n",
            "235/235 [==============================] - 1s 3ms/step - loss: 0.0598 - accuracy: 0.9807 - val_loss: 0.0763 - val_accuracy: 0.9773\n",
            "Epoch 13/50\n",
            "235/235 [==============================] - 1s 3ms/step - loss: 0.0585 - accuracy: 0.9814 - val_loss: 0.0816 - val_accuracy: 0.9751\n",
            "Epoch 14/50\n",
            "235/235 [==============================] - 1s 4ms/step - loss: 0.0582 - accuracy: 0.9817 - val_loss: 0.0811 - val_accuracy: 0.9764\n",
            "Epoch 15/50\n",
            "235/235 [==============================] - 1s 3ms/step - loss: 0.0540 - accuracy: 0.9836 - val_loss: 0.0737 - val_accuracy: 0.9776\n",
            "Epoch 16/50\n",
            "235/235 [==============================] - 1s 3ms/step - loss: 0.0554 - accuracy: 0.9824 - val_loss: 0.0769 - val_accuracy: 0.9776\n",
            "Epoch 17/50\n",
            "235/235 [==============================] - 1s 3ms/step - loss: 0.0543 - accuracy: 0.9826 - val_loss: 0.0781 - val_accuracy: 0.9755\n",
            "Epoch 18/50\n",
            "235/235 [==============================] - 1s 3ms/step - loss: 0.0551 - accuracy: 0.9823 - val_loss: 0.0840 - val_accuracy: 0.9757\n",
            "Epoch 19/50\n",
            "235/235 [==============================] - 1s 4ms/step - loss: 0.0532 - accuracy: 0.9833 - val_loss: 0.0861 - val_accuracy: 0.9718\n",
            "Epoch 20/50\n",
            "235/235 [==============================] - 1s 3ms/step - loss: 0.0530 - accuracy: 0.9829 - val_loss: 0.0774 - val_accuracy: 0.9763\n",
            "Epoch 21/50\n",
            "235/235 [==============================] - 1s 3ms/step - loss: 0.0530 - accuracy: 0.9831 - val_loss: 0.0824 - val_accuracy: 0.9753\n",
            "Epoch 22/50\n",
            "235/235 [==============================] - 1s 4ms/step - loss: 0.0505 - accuracy: 0.9840 - val_loss: 0.0759 - val_accuracy: 0.9777\n",
            "Epoch 23/50\n",
            "235/235 [==============================] - 1s 4ms/step - loss: 0.0496 - accuracy: 0.9844 - val_loss: 0.0790 - val_accuracy: 0.9776\n",
            "Epoch 24/50\n",
            "235/235 [==============================] - 1s 5ms/step - loss: 0.0494 - accuracy: 0.9837 - val_loss: 0.0771 - val_accuracy: 0.9770\n",
            "Epoch 25/50\n",
            "235/235 [==============================] - 1s 5ms/step - loss: 0.0474 - accuracy: 0.9849 - val_loss: 0.0800 - val_accuracy: 0.9754\n",
            "Epoch 26/50\n",
            "235/235 [==============================] - 1s 3ms/step - loss: 0.0466 - accuracy: 0.9848 - val_loss: 0.0776 - val_accuracy: 0.9760\n",
            "Epoch 27/50\n",
            "235/235 [==============================] - 1s 3ms/step - loss: 0.0471 - accuracy: 0.9852 - val_loss: 0.0748 - val_accuracy: 0.9764\n",
            "Epoch 28/50\n",
            "235/235 [==============================] - 1s 3ms/step - loss: 0.0454 - accuracy: 0.9857 - val_loss: 0.0797 - val_accuracy: 0.9757\n",
            "Epoch 29/50\n",
            "235/235 [==============================] - 1s 3ms/step - loss: 0.0461 - accuracy: 0.9855 - val_loss: 0.0783 - val_accuracy: 0.9757\n",
            "Epoch 30/50\n",
            "235/235 [==============================] - 1s 3ms/step - loss: 0.0441 - accuracy: 0.9857 - val_loss: 0.0774 - val_accuracy: 0.9764\n",
            "Epoch 31/50\n",
            "235/235 [==============================] - 1s 3ms/step - loss: 0.0440 - accuracy: 0.9854 - val_loss: 0.0750 - val_accuracy: 0.9781\n",
            "Epoch 32/50\n",
            "235/235 [==============================] - 1s 3ms/step - loss: 0.0438 - accuracy: 0.9859 - val_loss: 0.0791 - val_accuracy: 0.9769\n",
            "Epoch 33/50\n",
            "235/235 [==============================] - 1s 3ms/step - loss: 0.0423 - accuracy: 0.9864 - val_loss: 0.0790 - val_accuracy: 0.9772\n",
            "Epoch 34/50\n",
            "235/235 [==============================] - 1s 3ms/step - loss: 0.0420 - accuracy: 0.9868 - val_loss: 0.0792 - val_accuracy: 0.9765\n",
            "Epoch 35/50\n",
            "235/235 [==============================] - 1s 3ms/step - loss: 0.0409 - accuracy: 0.9865 - val_loss: 0.0846 - val_accuracy: 0.9746\n",
            "Epoch 36/50\n",
            "235/235 [==============================] - 1s 3ms/step - loss: 0.0400 - accuracy: 0.9869 - val_loss: 0.0774 - val_accuracy: 0.9775\n",
            "Epoch 37/50\n",
            "235/235 [==============================] - 1s 3ms/step - loss: 0.0403 - accuracy: 0.9867 - val_loss: 0.0784 - val_accuracy: 0.9768\n",
            "Epoch 38/50\n",
            "235/235 [==============================] - 1s 3ms/step - loss: 0.0399 - accuracy: 0.9866 - val_loss: 0.0831 - val_accuracy: 0.9750\n",
            "Epoch 39/50\n",
            "235/235 [==============================] - 1s 4ms/step - loss: 0.0396 - accuracy: 0.9870 - val_loss: 0.0875 - val_accuracy: 0.9750\n",
            "Epoch 40/50\n",
            "235/235 [==============================] - 1s 5ms/step - loss: 0.0413 - accuracy: 0.9867 - val_loss: 0.0822 - val_accuracy: 0.9756\n",
            "Epoch 41/50\n",
            "235/235 [==============================] - 1s 5ms/step - loss: 0.0390 - accuracy: 0.9875 - val_loss: 0.0795 - val_accuracy: 0.9767\n",
            "Epoch 42/50\n",
            "235/235 [==============================] - 1s 5ms/step - loss: 0.0394 - accuracy: 0.9874 - val_loss: 0.0813 - val_accuracy: 0.9781\n",
            "Epoch 43/50\n",
            "235/235 [==============================] - 1s 3ms/step - loss: 0.0383 - accuracy: 0.9874 - val_loss: 0.0829 - val_accuracy: 0.9766\n",
            "Epoch 44/50\n",
            "235/235 [==============================] - 1s 3ms/step - loss: 0.0373 - accuracy: 0.9876 - val_loss: 0.0824 - val_accuracy: 0.9763\n",
            "Epoch 45/50\n",
            "235/235 [==============================] - 1s 3ms/step - loss: 0.0376 - accuracy: 0.9874 - val_loss: 0.0868 - val_accuracy: 0.9751\n",
            "Epoch 46/50\n",
            "235/235 [==============================] - 1s 3ms/step - loss: 0.0366 - accuracy: 0.9880 - val_loss: 0.0869 - val_accuracy: 0.9736\n",
            "Epoch 47/50\n",
            "235/235 [==============================] - 1s 3ms/step - loss: 0.0372 - accuracy: 0.9874 - val_loss: 0.0896 - val_accuracy: 0.9739\n",
            "Epoch 48/50\n",
            "235/235 [==============================] - 1s 3ms/step - loss: 0.0350 - accuracy: 0.9890 - val_loss: 0.0904 - val_accuracy: 0.9738\n",
            "Epoch 49/50\n",
            "235/235 [==============================] - 1s 3ms/step - loss: 0.0353 - accuracy: 0.9880 - val_loss: 0.0794 - val_accuracy: 0.9782\n",
            "Epoch 50/50\n",
            "235/235 [==============================] - 1s 3ms/step - loss: 0.0332 - accuracy: 0.9890 - val_loss: 0.0848 - val_accuracy: 0.9759\n"
          ]
        }
      ],
      "source": [
        "# Convert labels to categorical\n",
        "y_train = to_categorical(train_labels)\n",
        "y_test = to_categorical(test_labels)\n",
        "\n",
        "# Compile the classifier model\n",
        "classifier.compile(optimizer='adam', loss='categorical_crossentropy', metrics=['accuracy'])\n",
        "\n",
        "# Train the classifier\n",
        "history = classifier.fit(x_train_encoded, y_train,\n",
        "                         epochs=50, batch_size=256, shuffle=True,\n",
        "                         validation_data=(x_test_encoded, y_test))"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {},
      "source": [
        "---"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {},
      "source": [
        "- Plot accuracy, accuracy validation, loss and loss validation"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 263
        },
        "id": "iP6TdVtDYnDJ",
        "outputId": "3b09e052-36ec-4a50-d212-7649c27cf427"
      },
      "outputs": [
        {
          "data": {
            "image/png": "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",
            "text/plain": [
              "<Figure size 1200x400 with 2 Axes>"
            ]
          },
          "metadata": {},
          "output_type": "display_data"
        }
      ],
      "source": [
        "plt.figure(figsize=(12, 4))\n",
        "plt.subplot(1, 2, 1)\n",
        "plt.plot(history.history['accuracy'])\n",
        "plt.plot(history.history['val_accuracy'])\n",
        "plt.title('Model accuracy')\n",
        "plt.ylabel('Accuracy')\n",
        "plt.xlabel('Epoch')\n",
        "plt.legend(['Train', 'Validation'], loc='upper left')\n",
        "\n",
        "plt.subplot(1, 2, 2)\n",
        "plt.plot(history.history['loss'])\n",
        "plt.plot(history.history['val_loss'])\n",
        "plt.title('Model loss')\n",
        "plt.ylabel('Loss')\n",
        "plt.xlabel('Epoch')\n",
        "plt.legend(['Train', 'Validation'], loc='upper left')\n",
        "plt.show()"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {},
      "source": [
        "- Report accuracy after training phase"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "2ZQVjOgnYnc5",
        "outputId": "72901ae3-cfe1-4004-848c-6dd59a89ac13"
      },
      "outputs": [
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "1875/1875 [==============================] - 4s 2ms/step - loss: 0.0331 - accuracy: 0.9892\n",
            "313/313 [==============================] - 1s 2ms/step - loss: 0.0848 - accuracy: 0.9759\n",
            "Accuracy: 0.9891999959945679\n",
            "Accuracy validation: 0.9758999943733215\n",
            "Loss: 0.03307820484042168\n",
            "Loss validation: 0.08484704792499542\n"
          ]
        }
      ],
      "source": [
        "train_loss, train_acc = classifier.evaluate(x_train_encoded, y_train)\n",
        "test_loss, test_acc = classifier.evaluate(x_test_encoded, y_test)\n",
        "print('Accuracy:', train_acc)\n",
        "print('Accuracy validation:', test_acc)\n",
        "\n",
        "print('Loss:', train_loss)\n",
        "print('Loss validation:', test_loss)"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {},
      "source": [
        "-  confusion matrix"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 490
        },
        "id": "RNPqkGAoBgV-",
        "outputId": "4abf9267-77cc-4f8f-db99-696ea33ea072"
      },
      "outputs": [
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "313/313 [==============================] - 1s 3ms/step\n"
          ]
        },
        {
          "data": {
            "image/png": "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",
            "text/plain": [
              "<Figure size 640x480 with 2 Axes>"
            ]
          },
          "metadata": {},
          "output_type": "display_data"
        }
      ],
      "source": [
        "y_pred = classifier.predict(x_test_encoded)\n",
        "y_pred_classes = np.argmax(y_pred, axis=1)\n",
        "y_true = np.argmax(y_test, axis=1)\n",
        "confusion_mtx = confusion_matrix(y_true, y_pred_classes)\n",
        "sns.heatmap(confusion_mtx, annot=True, fmt='d')\n",
        "plt.title(\"Confusion Matrix\")\n",
        "plt.ylabel('True label')\n",
        "plt.xlabel('Predicted label')\n",
        "plt.show()"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {},
      "source": [
        "As it shows, the model classify the validation data with high accuracy correctly. For example `937` pictures with label `6` are classify `correctly` and 21 pictures are misclassified."
      ]
    }
  ],
  "metadata": {
    "colab": {
      "provenance": []
    },
    "kernelspec": {
      "display_name": "Python 3",
      "name": "python3"
    },
    "language_info": {
      "name": "python"
    }
  },
  "nbformat": 4,
  "nbformat_minor": 0
}