NassimElH01/kopi-af-fashion_classification_neural_nets-ipynb-1b.ipynb

## kopi-af-fashion_classification_neural_nets-ipynb-1b.ipynb
{
  "cells": [
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "view-in-github",
        "colab_type": "text"
      },
      "source": [
        "<a href=\"https://colab.research.google.com/gist/NassimElH01/7432e41fc6a1d18f1982faca6d64263b/kopi-af-fashion_classification_neural_nets-ipynb-1b.ipynb\" target=\"_parent\"><img src=\"https://colab.research.google.com/assets/colab-badge.svg\" alt=\"Open In Colab\"/></a>"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "SHyYtLVakc06"
      },
      "source": [
        "# Setup"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "tuBXppANkc07"
      },
      "source": [
        "This project requires Python 3.7 or above:"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "id": "eIwILkKJkc08"
      },
      "outputs": [],
      "source": [
        "import sys\n",
        "\n",
        "assert sys.version_info >= (3, 7)"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "hDaPBZ0_kc1A"
      },
      "source": [
        "And TensorFlow ≥ 2.8:"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "id": "9lqbTqQAkc1B"
      },
      "outputs": [],
      "source": [
        "import tensorflow as tf\n",
        "from packaging import version\n",
        "assert version.parse(tf.__version__) >= version.parse(\"2.8.0\")"
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "import matplotlib.pyplot as plt\n",
        "\n",
        "plt.rc('font', size=14)\n",
        "plt.rc('axes', labelsize=14, titlesize=14)\n",
        "plt.rc('legend', fontsize=14)\n",
        "plt.rc('xtick', labelsize=10)\n",
        "plt.rc('ytick', labelsize=10)"
      ],
      "metadata": {
        "id": "EkhHQS7Y1I2O"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "import numpy as np"
      ],
      "metadata": {
        "id": "QJw59sFq18Qu"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "pR6kI53_kc1U"
      },
      "source": [
        "# Implementing MLPs with Keras\n",
        "## Building an Image Classifier Using the Sequential API\n",
        "### Using Keras to load the dataset"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "6_tQFrQYkc1U"
      },
      "source": [
        "Let's start by loading the fashion MNIST dataset. Keras has a number of functions to load popular datasets in `tf.keras.datasets`. The dataset is already split for you between a training set (60,000 images) and a test set (10,000 images), but it can be useful to split the training set further to have a validation set. We'll use 55,000 images for training, and 5,000 for validation."
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "id": "4sJ3RDoYkc1U",
        "outputId": "4d30aa7c-b4c2-4839-e0ea-606a523329c7",
        "colab": {
          "base_uri": "https://localhost:8080/"
        }
      },
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Downloading data from https://storage.googleapis.com/tensorflow/tf-keras-datasets/train-labels-idx1-ubyte.gz\n",
            "\u001b[1m29515/29515\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 0us/step\n",
            "Downloading data from https://storage.googleapis.com/tensorflow/tf-keras-datasets/train-images-idx3-ubyte.gz\n",
            "\u001b[1m26421880/26421880\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 0us/step\n",
            "Downloading data from https://storage.googleapis.com/tensorflow/tf-keras-datasets/t10k-labels-idx1-ubyte.gz\n",
            "\u001b[1m5148/5148\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 4us/step\n",
            "Downloading data from https://storage.googleapis.com/tensorflow/tf-keras-datasets/t10k-images-idx3-ubyte.gz\n",
            "\u001b[1m4422102/4422102\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 0us/step\n"
          ]
        }
      ],
      "source": [
        "import tensorflow as tf\n",
        "\n",
        "fashion_mnist = tf.keras.datasets.fashion_mnist.load_data()\n",
        "(X_train_full, y_train_full), (X_test, y_test) = fashion_mnist\n",
        "X_train, y_train = X_train_full[:-5000], y_train_full[:-5000]\n",
        "X_valid, y_valid = X_train_full[-5000:], y_train_full[-5000:]"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "-4YnMmNckc1V"
      },
      "source": [
        "The training set contains 60,000 grayscale images, each 28x28 pixels:"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "EgD8_yySkc1V",
        "outputId": "ddb769b1-fb1b-42fb-f4b3-70593dda5041"
      },
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "(55000, 28, 28)"
            ]
          },
          "metadata": {},
          "execution_count": 6
        }
      ],
      "source": [
        "X_train.shape"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "HI7KWMtwkc1W"
      },
      "source": [
        "Each pixel intensity is represented as a byte (0 to 255):"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "mqbvKfHlkc1W",
        "outputId": "3da8ffcc-e919-401b-fe05-db8162119631"
      },
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "dtype('uint8')"
            ]
          },
          "metadata": {},
          "execution_count": 7
        }
      ],
      "source": [
        "X_train.dtype"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "b08DmMkRkc1X"
      },
      "source": [
        "Let's scale the pixel intensities down to the 0-1 range and convert them to floats, by dividing by 255:"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "id": "lJfk6ryskc1X"
      },
      "outputs": [],
      "source": [
        "X_train, X_valid, X_test = X_train / 255., X_valid / 255., X_test / 255."
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "zhZF_2Kakc1X"
      },
      "source": [
        "You can plot an image using Matplotlib's `imshow()` function, with a `'binary'`\n",
        " color map:"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "OvDm8cL7kc1Z"
      },
      "source": [
        "The labels are the class IDs (represented as uint8), from 0 to 9:"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "tRlr8Ws5kc1a",
        "outputId": "2b750a8a-1bbb-4d36-b290-6c5904cfc32b"
      },
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "array([9, 0, 0, ..., 9, 0, 2], dtype=uint8)"
            ]
          },
          "metadata": {},
          "execution_count": 9
        }
      ],
      "source": [
        "y_train"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "rKk-W1lSkc1a"
      },
      "source": [
        "Here are the corresponding class names:"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "id": "bK2o6bHbkc1a"
      },
      "outputs": [],
      "source": [
        "class_names = [\"T-shirt/top\", \"Trouser\", \"Pullover\", \"Dress\", \"Coat\",\n",
        "               \"Sandal\", \"Shirt\", \"Sneaker\", \"Bag\", \"Ankle boot\"]"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "bMfIcgFskc1a"
      },
      "source": [
        "So the first image in the training set is an ankle boot:"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 37
        },
        "id": "x0L9dsSjkc1b",
        "outputId": "3984ab0a-7718-4426-da02-dc137866fd4a"
      },
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "'Ankle boot'"
            ],
            "application/vnd.google.colaboratory.intrinsic+json": {
              "type": "string"
            }
          },
          "metadata": {},
          "execution_count": 11
        }
      ],
      "source": [
        "class_names[y_train[0]]"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "TWWqJSvHkc1b"
      },
      "source": [
        "Let's take a look at a sample of the images in the dataset:"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 429
        },
        "id": "5w5kxaYpkc1b",
        "outputId": "31c32bae-3780-478c-fde6-6bc186a95ee8"
      },
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 1200x480 with 40 Axes>"
            ],
            "image/png": "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\n"
          },
          "metadata": {}
        }
      ],
      "source": [
        "# extra code – this cell generates and saves Figure 10–10\n",
        "\n",
        "n_rows = 4\n",
        "n_cols = 10\n",
        "plt.figure(figsize=(n_cols * 1.2, n_rows * 1.2))\n",
        "for row in range(n_rows):\n",
        "    for col in range(n_cols):\n",
        "        index = n_cols * row + col\n",
        "        plt.subplot(n_rows, n_cols, index + 1)\n",
        "        plt.imshow(X_train[index], cmap=\"binary\", interpolation=\"nearest\")\n",
        "        plt.axis('off')\n",
        "        plt.title(class_names[y_train[index]])\n",
        "plt.subplots_adjust(wspace=0.2, hspace=0.5)\n",
        "\n",
        "#save_fig(\"fashion_mnist_plot\")\n",
        "plt.show()"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "duf-qz1Xkc1c"
      },
      "source": [
        "### Creating the model using the Sequential API"
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "###Parameters for execises\n"
      ],
      "metadata": {
        "id": "J3dSvrV2OC_T"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "numberOfIterations=10\n",
        "#numberOfIterations=30"
      ],
      "metadata": {
        "id": "HYv9bZkzEilX"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "ActivationFunction = \"relu\"\n",
        "# ActivationFunction = \"sigmoid\"\n",
        "# ActivationFunction = \"tanh\""
      ],
      "metadata": {
        "id": "oCb1ZLySBJFi"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "Neurons1 = 300\n",
        "# Neurons1 = 50\n"
      ],
      "metadata": {
        "id": "ykst3XQBDeSZ"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "Neurons2 = 100\n",
        "# Neurons2 = 50"
      ],
      "metadata": {
        "id": "rAVJYx91DqB6"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "id": "08DaJeg3kc1c",
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "outputId": "ab7be220-09d4-4485-ca04-91569fbca8c0"
      },
      "outputs": [
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "/usr/local/lib/python3.10/dist-packages/keras/src/layers/core/input_layer.py:26: UserWarning: Argument `input_shape` is deprecated. Use `shape` instead.\n",
            "  warnings.warn(\n"
          ]
        }
      ],
      "source": [
        "tf.random.set_seed(42)\n",
        "model = tf.keras.Sequential()\n",
        "model.add(tf.keras.layers.InputLayer(input_shape=[28, 28]))\n",
        "model.add(tf.keras.layers.Flatten())\n",
        "model.add(tf.keras.layers.Dense(Neurons1, activation=ActivationFunction))\n",
        "model.add(tf.keras.layers.Dense(Neurons2, activation=ActivationFunction))\n",
        "model.add(tf.keras.layers.Dense(10, activation=\"softmax\"))"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "id": "DD0Sb1AVkc1d",
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "outputId": "6bd773eb-bd6e-438f-88b6-24bf8d3188dd"
      },
      "outputs": [
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "/usr/local/lib/python3.10/dist-packages/keras/src/layers/reshaping/flatten.py:37: UserWarning: Do not pass an `input_shape`/`input_dim` argument to a layer. When using Sequential models, prefer using an `Input(shape)` object as the first layer in the model instead.\n",
            "  super().__init__(**kwargs)\n"
          ]
        }
      ],
      "source": [
        "# extra code – clear the session to reset the name counters\n",
        "tf.keras.backend.clear_session()\n",
        "tf.random.set_seed(42)\n",
        "\n",
        "model = tf.keras.Sequential([\n",
        "    tf.keras.layers.Flatten(input_shape=[28, 28]),\n",
        "    tf.keras.layers.Dense(Neurons1, activation=ActivationFunction),\n",
        "    tf.keras.layers.Dense(Neurons2, activation=ActivationFunction),\n",
        "    tf.keras.layers.Dense(10, activation=\"softmax\")\n",
        "])"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 257
        },
        "id": "1eWhAT_Jkc1d",
        "outputId": "a52252e4-6a31-42d8-ab71-955f177d63ad"
      },
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "\u001b[1mModel: \"sequential\"\u001b[0m\n"
            ],
            "text/html": [
              "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"font-weight: bold\">Model: \"sequential\"</span>\n",
              "</pre>\n"
            ]
          },
          "metadata": {}
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "┏━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━┓\n",
              "┃\u001b[1m \u001b[0m\u001b[1mLayer (type)                        \u001b[0m\u001b[1m \u001b[0m┃\u001b[1m \u001b[0m\u001b[1mOutput Shape               \u001b[0m\u001b[1m \u001b[0m┃\u001b[1m \u001b[0m\u001b[1m        Param #\u001b[0m\u001b[1m \u001b[0m┃\n",
              "┡━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━┩\n",
              "│ flatten (\u001b[38;5;33mFlatten\u001b[0m)                    │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m784\u001b[0m)                 │               \u001b[38;5;34m0\u001b[0m │\n",
              "├──────────────────────────────────────┼─────────────────────────────┼─────────────────┤\n",
              "│ dense (\u001b[38;5;33mDense\u001b[0m)                        │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m300\u001b[0m)                 │         \u001b[38;5;34m235,500\u001b[0m │\n",
              "├──────────────────────────────────────┼─────────────────────────────┼─────────────────┤\n",
              "│ dense_1 (\u001b[38;5;33mDense\u001b[0m)                      │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m100\u001b[0m)                 │          \u001b[38;5;34m30,100\u001b[0m │\n",
              "├──────────────────────────────────────┼─────────────────────────────┼─────────────────┤\n",
              "│ dense_2 (\u001b[38;5;33mDense\u001b[0m)                      │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m10\u001b[0m)                  │           \u001b[38;5;34m1,010\u001b[0m │\n",
              "└──────────────────────────────────────┴─────────────────────────────┴─────────────────┘\n"
            ],
            "text/html": [
              "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\">┏━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━┓\n",
              "┃<span style=\"font-weight: bold\"> Layer (type)                         </span>┃<span style=\"font-weight: bold\"> Output Shape                </span>┃<span style=\"font-weight: bold\">         Param # </span>┃\n",
              "┡━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━┩\n",
              "│ flatten (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">Flatten</span>)                    │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">784</span>)                 │               <span style=\"color: #00af00; text-decoration-color: #00af00\">0</span> │\n",
              "├──────────────────────────────────────┼─────────────────────────────┼─────────────────┤\n",
              "│ dense (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">Dense</span>)                        │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">300</span>)                 │         <span style=\"color: #00af00; text-decoration-color: #00af00\">235,500</span> │\n",
              "├──────────────────────────────────────┼─────────────────────────────┼─────────────────┤\n",
              "│ dense_1 (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">Dense</span>)                      │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">100</span>)                 │          <span style=\"color: #00af00; text-decoration-color: #00af00\">30,100</span> │\n",
              "├──────────────────────────────────────┼─────────────────────────────┼─────────────────┤\n",
              "│ dense_2 (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">Dense</span>)                      │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">10</span>)                  │           <span style=\"color: #00af00; text-decoration-color: #00af00\">1,010</span> │\n",
              "└──────────────────────────────────────┴─────────────────────────────┴─────────────────┘\n",
              "</pre>\n"
            ]
          },
          "metadata": {}
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "\u001b[1m Total params: \u001b[0m\u001b[38;5;34m266,610\u001b[0m (1.02 MB)\n"
            ],
            "text/html": [
              "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"font-weight: bold\"> Total params: </span><span style=\"color: #00af00; text-decoration-color: #00af00\">266,610</span> (1.02 MB)\n",
              "</pre>\n"
            ]
          },
          "metadata": {}
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "\u001b[1m Trainable params: \u001b[0m\u001b[38;5;34m266,610\u001b[0m (1.02 MB)\n"
            ],
            "text/html": [
              "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"font-weight: bold\"> Trainable params: </span><span style=\"color: #00af00; text-decoration-color: #00af00\">266,610</span> (1.02 MB)\n",
              "</pre>\n"
            ]
          },
          "metadata": {}
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "\u001b[1m Non-trainable params: \u001b[0m\u001b[38;5;34m0\u001b[0m (0.00 B)\n"
            ],
            "text/html": [
              "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"font-weight: bold\"> Non-trainable params: </span><span style=\"color: #00af00; text-decoration-color: #00af00\">0</span> (0.00 B)\n",
              "</pre>\n"
            ]
          },
          "metadata": {}
        }
      ],
      "source": [
        "model.summary()"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 1000
        },
        "id": "dfrH5gy5kc1e",
        "outputId": "f2c4667e-c974-4851-b524-e697ede3f739"
      },
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<IPython.core.display.Image object>"
            ]
          },
          "metadata": {},
          "execution_count": 20
        }
      ],
      "source": [
        "# extra code – another way to display the model's architecture\n",
        "tf.keras.utils.plot_model(model, \"my_fashion_mnist_model.png\", show_shapes=True)"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "Z_xwV8bVkc1e",
        "outputId": "10f674fa-27c7-44b3-d2e3-1e20f23f608a"
      },
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "[<Flatten name=flatten, built=True>,\n",
              " <Dense name=dense, built=True>,\n",
              " <Dense name=dense_1, built=True>,\n",
              " <Dense name=dense_2, built=True>]"
            ]
          },
          "metadata": {},
          "execution_count": 21
        }
      ],
      "source": [
        "model.layers"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 37
        },
        "id": "6uIMiKOokc1f",
        "outputId": "93b186ec-0cd7-41da-9255-8a9b6d27c84f"
      },
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "'dense'"
            ],
            "application/vnd.google.colaboratory.intrinsic+json": {
              "type": "string"
            }
          },
          "metadata": {},
          "execution_count": 22
        }
      ],
      "source": [
        "hidden1 = model.layers[1]\n",
        "hidden1.name"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "52oeEoCAkc1f",
        "outputId": "3dab7150-dfc4-489c-f7bb-dc2c38e824a3"
      },
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "True"
            ]
          },
          "metadata": {},
          "execution_count": 23
        }
      ],
      "source": [
        "model.get_layer('dense') is hidden1"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "m-JvaxZokc1f",
        "outputId": "f0f0e59a-e03a-4d58-89ca-4af8b62082fa"
      },
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "array([[ 0.07221992, -0.06667063,  0.02500913, ...,  0.03793097,\n",
              "        -0.00233526, -0.06408731],\n",
              "       [-0.03879976, -0.01351639, -0.03285287, ..., -0.02183819,\n",
              "        -0.06373513, -0.06718601],\n",
              "       [ 0.00740068, -0.02750799,  0.06750055, ..., -0.03917148,\n",
              "         0.00631178, -0.02407178],\n",
              "       ...,\n",
              "       [-0.02778303, -0.0493311 ,  0.02145544, ...,  0.04771965,\n",
              "         0.03535071, -0.05551457],\n",
              "       [ 0.06495601,  0.02056506, -0.01449646, ...,  0.02841979,\n",
              "         0.03264547, -0.00527623],\n",
              "       [ 0.06434447, -0.02624152, -0.04732826, ..., -0.00567847,\n",
              "         0.0472416 ,  0.02259164]], dtype=float32)"
            ]
          },
          "metadata": {},
          "execution_count": 24
        }
      ],
      "source": [
        "weights, biases = hidden1.get_weights()\n",
        "weights"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "41PEVev9kc1g",
        "outputId": "0f6f5612-2fec-4883-fc65-161d8680fe08"
      },
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "(784, 300)"
            ]
          },
          "metadata": {},
          "execution_count": 25
        }
      ],
      "source": [
        "weights.shape"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "izdZvwVwkc1g",
        "outputId": "a2b319e2-a804-4f43-d7b5-cd386835a7e2"
      },
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "array([0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0.,\n",
              "       0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0.,\n",
              "       0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0.,\n",
              "       0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0.,\n",
              "       0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0.,\n",
              "       0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0.,\n",
              "       0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0.,\n",
              "       0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0.,\n",
              "       0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0.,\n",
              "       0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0.,\n",
              "       0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0.,\n",
              "       0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0.,\n",
              "       0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0.,\n",
              "       0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0.,\n",
              "       0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0.,\n",
              "       0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0.,\n",
              "       0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0.,\n",
              "       0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0.], dtype=float32)"
            ]
          },
          "metadata": {},
          "execution_count": 26
        }
      ],
      "source": [
        "biases"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "5IGJwHXLkc1g",
        "outputId": "e659294c-de3b-418b-deb1-f0af912387a6"
      },
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "(300,)"
            ]
          },
          "metadata": {},
          "execution_count": 27
        }
      ],
      "source": [
        "biases.shape"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "WZehB6t4kc1h"
      },
      "source": [
        "### Compiling the model"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "id": "PyQ9bs61kc1h"
      },
      "outputs": [],
      "source": [
        "model.compile(loss=\"sparse_categorical_crossentropy\",\n",
        "              optimizer=\"sgd\",\n",
        "              metrics=[\"accuracy\"])"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "7i0SIs6-kc1i"
      },
      "source": [
        "This is equivalent to:"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "id": "J6DmHgJWkc1i"
      },
      "outputs": [],
      "source": [
        "# extra code – this cell is equivalent to the previous cell\n",
        "model.compile(loss=tf.keras.losses.sparse_categorical_crossentropy,\n",
        "              optimizer=tf.keras.optimizers.SGD(),\n",
        "              metrics=[tf.keras.metrics.sparse_categorical_accuracy])"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "-dq1hepYkc1i",
        "outputId": "5bbdc440-478d-4a73-e836-1a059ee9d6e0"
      },
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "array([[1., 0., 0., 0., 0., 0., 0., 0., 0., 0.],\n",
              "       [0., 0., 0., 0., 0., 1., 0., 0., 0., 0.],\n",
              "       [0., 1., 0., 0., 0., 0., 0., 0., 0., 0.],\n",
              "       [1., 0., 0., 0., 0., 0., 0., 0., 0., 0.]])"
            ]
          },
          "metadata": {},
          "execution_count": 30
        }
      ],
      "source": [
        "# extra code – shows how to convert class ids to one-hot vectors\n",
        "tf.keras.utils.to_categorical([0, 5, 1, 0], num_classes=10)"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "nRKM0aaLkc1i"
      },
      "source": [
        "Note: it's important to set `num_classes` when the number of classes is greater than the maximum class id in the sample."
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "rgNSniUGkc1j",
        "outputId": "ff248cf0-3315-4d90-b088-e340a08f00fb"
      },
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "array([0, 5, 1, 0])"
            ]
          },
          "metadata": {},
          "execution_count": 31
        }
      ],
      "source": [
        "# extra code – shows how to convert one-hot vectors to class ids\n",
        "np.argmax(\n",
        "    [[1., 0., 0., 0., 0., 0., 0., 0., 0., 0.],\n",
        "     [0., 0., 0., 0., 0., 1., 0., 0., 0., 0.],\n",
        "     [0., 1., 0., 0., 0., 0., 0., 0., 0., 0.],\n",
        "     [1., 0., 0., 0., 0., 0., 0., 0., 0., 0.]],\n",
        "    axis=1\n",
        ")"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "A2tJ1UsVkc1j"
      },
      "source": [
        "### Training and evaluating the model"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "LiRFxXrmkc1j",
        "outputId": "1e228c8f-0882-4d80-e285-e7ff3d9a97e2"
      },
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Epoch 1/10\n",
            "\u001b[1m1719/1719\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m7s\u001b[0m 4ms/step - loss: 0.9986 - sparse_categorical_accuracy: 0.6814 - val_loss: 0.5106 - val_sparse_categorical_accuracy: 0.8286\n",
            "Epoch 2/10\n",
            "\u001b[1m1719/1719\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m14s\u001b[0m 6ms/step - loss: 0.5085 - sparse_categorical_accuracy: 0.8249 - val_loss: 0.4575 - val_sparse_categorical_accuracy: 0.8360\n",
            "Epoch 3/10\n",
            "\u001b[1m1719/1719\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m10s\u001b[0m 6ms/step - loss: 0.4538 - sparse_categorical_accuracy: 0.8437 - val_loss: 0.4320 - val_sparse_categorical_accuracy: 0.8434\n",
            "Epoch 4/10\n",
            "\u001b[1m1719/1719\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m8s\u001b[0m 5ms/step - loss: 0.4235 - sparse_categorical_accuracy: 0.8534 - val_loss: 0.4170 - val_sparse_categorical_accuracy: 0.8474\n",
            "Epoch 5/10\n",
            "\u001b[1m1719/1719\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m7s\u001b[0m 4ms/step - loss: 0.4020 - sparse_categorical_accuracy: 0.8610 - val_loss: 0.4046 - val_sparse_categorical_accuracy: 0.8502\n",
            "Epoch 6/10\n",
            "\u001b[1m1719/1719\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m10s\u001b[0m 4ms/step - loss: 0.3849 - sparse_categorical_accuracy: 0.8661 - val_loss: 0.3960 - val_sparse_categorical_accuracy: 0.8548\n",
            "Epoch 7/10\n",
            "\u001b[1m1719/1719\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m9s\u001b[0m 3ms/step - loss: 0.3708 - sparse_categorical_accuracy: 0.8699 - val_loss: 0.3899 - val_sparse_categorical_accuracy: 0.8562\n",
            "Epoch 8/10\n",
            "\u001b[1m1719/1719\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m7s\u001b[0m 4ms/step - loss: 0.3586 - sparse_categorical_accuracy: 0.8741 - val_loss: 0.3826 - val_sparse_categorical_accuracy: 0.8612\n",
            "Epoch 9/10\n",
            "\u001b[1m1719/1719\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m7s\u001b[0m 4ms/step - loss: 0.3476 - sparse_categorical_accuracy: 0.8773 - val_loss: 0.3750 - val_sparse_categorical_accuracy: 0.8636\n",
            "Epoch 10/10\n",
            "\u001b[1m1719/1719\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m7s\u001b[0m 4ms/step - loss: 0.3373 - sparse_categorical_accuracy: 0.8811 - val_loss: 0.3666 - val_sparse_categorical_accuracy: 0.8650\n"
          ]
        }
      ],
      "source": [
        "history = model.fit(X_train, y_train, epochs=numberOfIterations,\n",
        "                    validation_data=(X_valid, y_valid))"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "VSmm0l4ekc1k",
        "outputId": "15d838e5-29a0-405b-8cf7-3b55ce69912f"
      },
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "{'verbose': 'auto', 'epochs': 10, 'steps': 1719}"
            ]
          },
          "metadata": {},
          "execution_count": 33
        }
      ],
      "source": [
        "history.params"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "lmPdyfl4kc1k",
        "outputId": "3d3e0aa1-565d-4f4e-c4f9-caabe701d35c"
      },
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "[0, 1, 2, 3, 4, 5, 6, 7, 8, 9]\n"
          ]
        }
      ],
      "source": [
        "print(history.epoch)"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 475
        },
        "id": "4opDEJxKkc1l",
        "outputId": "4d7db554-7a1b-4ff5-b672-54e3e4db2526"
      },
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 800x500 with 1 Axes>"
            ],
            "image/png": "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\n"
          },
          "metadata": {}
        }
      ],
      "source": [
        "import matplotlib.pyplot as plt\n",
        "import pandas as pd\n",
        "\n",
        "pd.DataFrame(history.history).plot(\n",
        "    figsize=(8, 5), xlim=[0, 29], ylim=[0, 1], grid=True, xlabel=\"Epoch\",\n",
        "    style=[\"r--\", \"r--.\", \"b-\", \"b-*\"])\n",
        "plt.legend(loc=\"lower left\")  # extra code\n",
        "#save_fig(\"keras_learning_curves_plot\")  # extra code\n",
        "plt.show()"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 475
        },
        "id": "AA9MA3vXkc1l",
        "outputId": "cd02aa95-8195-42ef-bb11-2c518d9e2ea2"
      },
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 800x500 with 1 Axes>"
            ],
            "image/png": "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\n"
          },
          "metadata": {}
        }
      ],
      "source": [
        "# extra code – shows how to shift the training curve by -1/2 epoch\n",
        "plt.figure(figsize=(8, 5))\n",
        "for key, style in zip(history.history, [\"r--\", \"r--.\", \"b-\", \"b-*\"]):\n",
        "    epochs = np.array(history.epoch) + (0 if key.startswith(\"val_\") else -0.5)\n",
        "    plt.plot(epochs, history.history[key], style, label=key)\n",
        "plt.xlabel(\"Epoch\")\n",
        "plt.axis([-0.5, 29, 0., 1])\n",
        "plt.legend(loc=\"lower left\")\n",
        "plt.grid()\n",
        "plt.show()"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "oDT8owWrkc1l",
        "outputId": "fdf272f0-83bf-4dde-eef9-28507799b8df"
      },
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "\u001b[1m313/313\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 2ms/step - loss: 0.3880 - sparse_categorical_accuracy: 0.8652\n"
          ]
        },
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "[0.39188262820243835, 0.861299991607666]"
            ]
          },
          "metadata": {},
          "execution_count": 37
        }
      ],
      "source": [
        "model.evaluate(X_test, y_test)"
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "###Excercise(s)"
      ],
      "metadata": {
        "id": "ry_aVCoXH-LJ"
      }
    },
    {
      "cell_type": "markdown",
      "source": [
        "###Exercise 1.\n",
        "Limit the number of neurons in both inner layers to 50 and numberOfIterations to 10.\n",
        "Note the accuraracy when changing between the following activation functions:\n",
        "\n",
        "relu:\n",
        "\n",
        "sigmoid:\n",
        "\n",
        "tanh:"
      ],
      "metadata": {
        "id": "N6MNQzNHGRKC"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "# prompt: Limit the number of neurons in both inner layers to 50 and numberOfIterations to 10. Note the accuraracy when changing between the following activation functions:\n",
        "# relu:\n",
        "# sigmoid:\n",
        "# tanh:\n",
        "\n",
        "import sys\n",
        "import tensorflow as tf\n",
        "from packaging import version\n",
        "import matplotlib.pyplot as plt\n",
        "import numpy as np\n",
        "import pandas as pd\n",
        "# # Setup\n",
        "# This project requires Python 3.7 or above:\n",
        "\n",
        "assert sys.version_info >= (3, 7)\n",
        "# And TensorFlow ≥ 2.8:\n",
        "assert version.parse(tf.__version__) >= version.parse(\"2.8.0\")\n",
        "\n",
        "plt.rc('font', size=14)\n",
        "plt.rc('axes', labelsize=14, titlesize=14)\n",
        "plt.rc('legend', fontsize=14)\n",
        "plt.rc('xtick', labelsize=10)\n",
        "plt.rc('ytick', labelsize=10)\n",
        "# # Implementing MLPs with Keras\n",
        "# ## Building an Image Classifier Using the Sequential API\n",
        "# ### Using Keras to load the dataset\n",
        "# Let's start by loading the fashion MNIST dataset. Keras has a number of functions to load popular datasets in `tf.keras.datasets`. The dataset is already split for you between a training set (60,000 images) and a test set (10,000 images), but it can be useful to split the training set further to have a validation set. We'll use 55,000 images for training, and 5,000 for validation.\n",
        "\n",
        "fashion_mnist = tf.keras.datasets.fashion_mnist.load_data()\n",
        "(X_train_full, y_train_full), (X_test, y_test) = fashion_mnist\n",
        "X_train, y_train = X_train_full[:-5000], y_train_full[:-5000]\n",
        "X_valid, y_valid = X_train_full[-5000:], y_train_full[-5000:]\n",
        "# The training set contains 60,000 grayscale images, each 28x28 pixels:\n",
        "X_train.shape\n",
        "# Each pixel intensity is represented as a byte (0 to 255):\n",
        "X_train.dtype\n",
        "# Let's scale the pixel intensities down to the 0-1 range and convert them to floats, by dividing by 255:\n",
        "X_train, X_valid, X_test = X_train / 255., X_valid / 255., X_test / 255.\n",
        "# You can plot an image using Matplotlib's `imshow()` function, with a `'binary'`\n",
        "#  color map:\n",
        "# The labels are the class IDs (represented as uint8), from 0 to 9:\n",
        "y_train\n",
        "# Here are the corresponding class names:\n",
        "class_names = [\"T-shirt/top\", \"Trouser\", \"Pullover\", \"Dress\", \"Coat\",\n",
        "               \"Sandal\", \"Shirt\", \"Sneaker\", \"Bag\", \"Ankle boot\"]\n",
        "# So the first image in the training set is an ankle boot:\n",
        "class_names[y_train[0]]\n",
        "# Let's take a look at a sample of the images in the dataset:\n",
        "# extra code – this cell generates and saves Figure 10–10\n",
        "\n",
        "n_rows = 4\n",
        "n_cols = 10\n",
        "plt.figure(figsize=(n_cols * 1.2, n_rows * 1.2))\n",
        "for row in range(n_rows):\n",
        "    for col in range(n_cols):\n",
        "        index = n_cols * row + col\n",
        "        plt.subplot(n_rows, n_cols, index + 1)\n",
        "        plt.imshow(X_train[index], cmap=\"binary\", interpolation=\"nearest\")\n",
        "        plt.axis('off')\n",
        "        plt.title(class_names[y_train[index]])\n",
        "plt.subplots_adjust(wspace=0.2, hspace=0.5)\n",
        "\n",
        "#save_fig(\"fashion_mnist_plot\")\n",
        "plt.show()\n",
        "# ### Creating the model using the Sequential API\n",
        "# ###Parameters for execises\n",
        "#\n",
        "numberOfIterations=10\n",
        "#numberOfIterations=30\n",
        "#ActivationFunction = \"relu\"\n",
        "# ActivationFunction = \"sigmoid\"\n",
        "# ActivationFunction = \"tanh\"\n",
        "#Neurons1 = 300\n",
        "Neurons1 = 50\n",
        "\n",
        "#Neurons2 = 100\n",
        "Neurons2 = 50\n",
        "\n",
        "\n",
        "activation_functions = [\"relu\", \"sigmoid\", \"tanh\"]\n",
        "\n",
        "for activation_function in activation_functions:\n",
        "  tf.random.set_seed(42)\n",
        "  model = tf.keras.Sequential()\n",
        "  model.add(tf.keras.layers.InputLayer(input_shape=[28, 28]))\n",
        "  model.add(tf.keras.layers.Flatten())\n",
        "  model.add(tf.keras.layers.Dense(Neurons1, activation=activation_function))\n",
        "  model.add(tf.keras.layers.Dense(Neurons2, activation=activation_function))\n",
        "  model.add(tf.keras.layers.Dense(10, activation=\"softmax\"))\n",
        "\n",
        "  model.compile(loss=\"sparse_categorical_crossentropy\",\n",
        "                optimizer=\"sgd\",\n",
        "                metrics=[\"accuracy\"])\n",
        "\n",
        "\n",
        "  history = model.fit(X_train, y_train, epochs=numberOfIterations,\n",
        "                      validation_data=(X_valid, y_valid))\n",
        "  _, accuracy = model.evaluate(X_test, y_test)\n",
        "  print(f\"Activation Function: {activation_function}, Accuracy: {accuracy}\")"
      ],
      "metadata": {
        "id": "P_YMRXbe9Z1t",
        "outputId": "dacad5df-ceda-458c-a919-9be13bb33b3b",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 1000
        }
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 1200x480 with 40 Axes>"
            ],
            "image/png": "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\n"
          },
          "metadata": {}
        },
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "/usr/local/lib/python3.10/dist-packages/keras/src/layers/core/input_layer.py:26: UserWarning: Argument `input_shape` is deprecated. Use `shape` instead.\n",
            "  warnings.warn(\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Epoch 1/10\n",
            "\u001b[1m1719/1719\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m4s\u001b[0m 2ms/step - accuracy: 0.6072 - loss: 1.1706 - val_accuracy: 0.8146 - val_loss: 0.5390\n",
            "Epoch 2/10\n",
            "\u001b[1m1719/1719\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 2ms/step - accuracy: 0.8130 - loss: 0.5391 - val_accuracy: 0.8306 - val_loss: 0.4762\n",
            "Epoch 3/10\n",
            "\u001b[1m1719/1719\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 2ms/step - accuracy: 0.8316 - loss: 0.4803 - val_accuracy: 0.8372 - val_loss: 0.4498\n",
            "Epoch 4/10\n",
            "\u001b[1m1719/1719\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m4s\u001b[0m 2ms/step - accuracy: 0.8429 - loss: 0.4505 - val_accuracy: 0.8430 - val_loss: 0.4334\n",
            "Epoch 5/10\n",
            "\u001b[1m1719/1719\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 2ms/step - accuracy: 0.8497 - loss: 0.4300 - val_accuracy: 0.8470 - val_loss: 0.4210\n",
            "Epoch 6/10\n",
            "\u001b[1m1719/1719\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 2ms/step - accuracy: 0.8554 - loss: 0.4143 - val_accuracy: 0.8504 - val_loss: 0.4119\n",
            "Epoch 7/10\n",
            "\u001b[1m1719/1719\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m6s\u001b[0m 2ms/step - accuracy: 0.8591 - loss: 0.4016 - val_accuracy: 0.8528 - val_loss: 0.4046\n",
            "Epoch 8/10\n",
            "\u001b[1m1719/1719\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m5s\u001b[0m 2ms/step - accuracy: 0.8627 - loss: 0.3907 - val_accuracy: 0.8538 - val_loss: 0.3993\n",
            "Epoch 9/10\n",
            "\u001b[1m1719/1719\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m5s\u001b[0m 2ms/step - accuracy: 0.8660 - loss: 0.3811 - val_accuracy: 0.8572 - val_loss: 0.3913\n",
            "Epoch 10/10\n",
            "\u001b[1m1719/1719\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m6s\u001b[0m 2ms/step - accuracy: 0.8694 - loss: 0.3721 - val_accuracy: 0.8588 - val_loss: 0.3848\n",
            "\u001b[1m313/313\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 1ms/step - accuracy: 0.8553 - loss: 0.4155\n",
            "Activation Function: relu, Accuracy: 0.8507999777793884\n",
            "Epoch 1/10\n",
            "\u001b[1m1719/1719\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m4s\u001b[0m 2ms/step - accuracy: 0.2806 - loss: 2.2573 - val_accuracy: 0.5386 - val_loss: 1.9187\n",
            "Epoch 2/10\n",
            "\u001b[1m1719/1719\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m4s\u001b[0m 2ms/step - accuracy: 0.5117 - loss: 1.7647 - val_accuracy: 0.5986 - val_loss: 1.3904\n",
            "Epoch 3/10\n",
            "\u001b[1m1719/1719\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m4s\u001b[0m 2ms/step - accuracy: 0.5960 - loss: 1.3114 - val_accuracy: 0.6530 - val_loss: 1.0950\n",
            "Epoch 4/10\n",
            "\u001b[1m1719/1719\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m4s\u001b[0m 2ms/step - accuracy: 0.6536 - loss: 1.0603 - val_accuracy: 0.7064 - val_loss: 0.9283\n",
            "Epoch 5/10\n",
            "\u001b[1m1719/1719\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m6s\u001b[0m 2ms/step - accuracy: 0.6969 - loss: 0.9158 - val_accuracy: 0.7414 - val_loss: 0.8229\n",
            "Epoch 6/10\n",
            "\u001b[1m1719/1719\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m5s\u001b[0m 3ms/step - accuracy: 0.7236 - loss: 0.8206 - val_accuracy: 0.7552 - val_loss: 0.7492\n",
            "Epoch 7/10\n",
            "\u001b[1m1719/1719\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m9s\u001b[0m 3ms/step - accuracy: 0.7425 - loss: 0.7526 - val_accuracy: 0.7656 - val_loss: 0.6952\n",
            "Epoch 8/10\n",
            "\u001b[1m1719/1719\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m4s\u001b[0m 2ms/step - accuracy: 0.7567 - loss: 0.7018 - val_accuracy: 0.7764 - val_loss: 0.6540\n",
            "Epoch 9/10\n",
            "\u001b[1m1719/1719\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 2ms/step - accuracy: 0.7668 - loss: 0.6626 - val_accuracy: 0.7802 - val_loss: 0.6216\n",
            "Epoch 10/10\n",
            "\u001b[1m1719/1719\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 2ms/step - accuracy: 0.7744 - loss: 0.6313 - val_accuracy: 0.7882 - val_loss: 0.5954\n",
            "\u001b[1m313/313\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 2ms/step - accuracy: 0.7773 - loss: 0.6252\n",
            "Activation Function: sigmoid, Accuracy: 0.7736999988555908\n",
            "Epoch 1/10\n",
            "\u001b[1m1719/1719\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m4s\u001b[0m 2ms/step - accuracy: 0.6460 - loss: 1.1299 - val_accuracy: 0.8140 - val_loss: 0.5430\n",
            "Epoch 2/10\n",
            "\u001b[1m1719/1719\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 2ms/step - accuracy: 0.8189 - loss: 0.5247 - val_accuracy: 0.8304 - val_loss: 0.4727\n",
            "Epoch 3/10\n",
            "\u001b[1m1719/1719\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m6s\u001b[0m 2ms/step - accuracy: 0.8377 - loss: 0.4617 - val_accuracy: 0.8392 - val_loss: 0.4430\n",
            "Epoch 4/10\n",
            "\u001b[1m1719/1719\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m4s\u001b[0m 2ms/step - accuracy: 0.8482 - loss: 0.4305 - val_accuracy: 0.8456 - val_loss: 0.4247\n",
            "Epoch 5/10\n",
            "\u001b[1m1719/1719\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m4s\u001b[0m 2ms/step - accuracy: 0.8546 - loss: 0.4101 - val_accuracy: 0.8506 - val_loss: 0.4115\n",
            "Epoch 6/10\n",
            "\u001b[1m1719/1719\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m6s\u001b[0m 2ms/step - accuracy: 0.8600 - loss: 0.3951 - val_accuracy: 0.8536 - val_loss: 0.4014\n",
            "Epoch 7/10\n",
            "\u001b[1m1719/1719\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 2ms/step - accuracy: 0.8646 - loss: 0.3833 - val_accuracy: 0.8560 - val_loss: 0.3934\n",
            "Epoch 8/10\n",
            "\u001b[1m1719/1719\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m5s\u001b[0m 2ms/step - accuracy: 0.8674 - loss: 0.3735 - val_accuracy: 0.8586 - val_loss: 0.3868\n",
            "Epoch 9/10\n",
            "\u001b[1m1719/1719\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m6s\u001b[0m 3ms/step - accuracy: 0.8701 - loss: 0.3651 - val_accuracy: 0.8602 - val_loss: 0.3814\n",
            "Epoch 10/10\n",
            "\u001b[1m1719/1719\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m4s\u001b[0m 2ms/step - accuracy: 0.8728 - loss: 0.3577 - val_accuracy: 0.8614 - val_loss: 0.3768\n",
            "\u001b[1m313/313\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 2ms/step - accuracy: 0.8599 - loss: 0.4053\n",
            "Activation Function: tanh, Accuracy: 0.8550000190734863\n"
          ]
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "###Exercise 2.\n",
        "\n",
        "Choose the best activation function.\n",
        "\n",
        "Try with different number neurons in the layers. Note the accuraracy when changing between the following combinations:\n",
        "\n",
        "Neurons1/Neurons2:\n",
        "\n",
        "300/300:\n",
        "\n",
        "200/200:\n",
        "\n",
        "100/100:\n",
        "\n",
        "300/100:\n",
        "\n",
        "100/300:"
      ],
      "metadata": {
        "id": "iT23UInEJDjd"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "# prompt: Choose the best activation function.\n",
        "# Try with different number neurons in the layers. Note the accuraracy when changing between the following combinations:\n",
        "# Neurons1/Neurons2:\n",
        "# 300/300:\n",
        "# 200/200:\n",
        "# 100/100:\n",
        "# 300/100:\n",
        "# 100/300:\n",
        "\n",
        "# ###Exercise 2.\n",
        "# Choose the best activation function.\n",
        "# Try with different number neurons in the layers. Note the accuraracy when changing between the following combinations:\n",
        "# Neurons1/Neurons2:\n",
        "# 300/300:\n",
        "# 200/200:\n",
        "# 100/100:\n",
        "# 300/100:\n",
        "# 100/300:\n",
        "\n",
        "\n",
        "numberOfIterations = 10\n",
        "activation_function = \"relu\"  # Choose the best activation function from Exercise 1\n",
        "\n",
        "neurons_combinations = [\n",
        "    (300, 300),\n",
        "    (200, 200),\n",
        "    (100, 100),\n",
        "    (300, 100),\n",
        "    (100, 300)\n",
        "]\n",
        "\n",
        "for neurons1, neurons2 in neurons_combinations:\n",
        "  tf.random.set_seed(42)\n",
        "  model = tf.keras.Sequential([\n",
        "      tf.keras.layers.Flatten(input_shape=[28, 28]),\n",
        "      tf.keras.layers.Dense(neurons1, activation=activation_function),\n",
        "      tf.keras.layers.Dense(neurons2, activation=activation_function),\n",
        "      tf.keras.layers.Dense(10, activation=\"softmax\")\n",
        "  ])\n",
        "\n",
        "  model.compile(loss=\"sparse_categorical_crossentropy\",\n",
        "                optimizer=\"sgd\",\n",
        "                metrics=[\"accuracy\"])\n",
        "\n",
        "  history = model.fit(X_train, y_train, epochs=numberOfIterations,\n",
        "                      validation_data=(X_valid, y_valid))\n",
        "  _, accuracy = model.evaluate(X_test, y_test)\n",
        "  print(f\"Neurons: {neurons1}/{neurons2}, Accuracy: {accuracy}\")"
      ],
      "metadata": {
        "id": "QqgTv5hl-usH",
        "outputId": "3133442e-e1e1-438c-9f08-bd8db59af1a4",
        "colab": {
          "base_uri": "https://localhost:8080/"
        }
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "/usr/local/lib/python3.10/dist-packages/keras/src/layers/reshaping/flatten.py:37: UserWarning: Do not pass an `input_shape`/`input_dim` argument to a layer. When using Sequential models, prefer using an `Input(shape)` object as the first layer in the model instead.\n",
            "  super().__init__(**kwargs)\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Epoch 1/10\n",
            "\u001b[1m1719/1719\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m13s\u001b[0m 7ms/step - accuracy: 0.6901 - loss: 0.9996 - val_accuracy: 0.8312 - val_loss: 0.5018\n",
            "Epoch 2/10\n",
            "\u001b[1m1719/1719\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m8s\u001b[0m 5ms/step - accuracy: 0.8268 - loss: 0.5045 - val_accuracy: 0.8424 - val_loss: 0.4522\n",
            "Epoch 3/10\n",
            "\u001b[1m1719/1719\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m8s\u001b[0m 4ms/step - accuracy: 0.8447 - loss: 0.4520 - val_accuracy: 0.8498 - val_loss: 0.4270\n",
            "Epoch 4/10\n",
            "\u001b[1m1719/1719\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m10s\u001b[0m 5ms/step - accuracy: 0.8530 - loss: 0.4218 - val_accuracy: 0.8544 - val_loss: 0.4112\n",
            "Epoch 5/10\n",
            "\u001b[1m1719/1719\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m6s\u001b[0m 4ms/step - accuracy: 0.8596 - loss: 0.4003 - val_accuracy: 0.8578 - val_loss: 0.4002\n",
            "Epoch 6/10\n",
            "\u001b[1m1719/1719\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m10s\u001b[0m 4ms/step - accuracy: 0.8660 - loss: 0.3833 - val_accuracy: 0.8580 - val_loss: 0.3914\n",
            "Epoch 7/10\n",
            "\u001b[1m1719/1719\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m11s\u001b[0m 4ms/step - accuracy: 0.8709 - loss: 0.3691 - val_accuracy: 0.8600 - val_loss: 0.3824\n",
            "Epoch 8/10\n",
            "\u001b[1m1719/1719\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m8s\u001b[0m 5ms/step - accuracy: 0.8747 - loss: 0.3568 - val_accuracy: 0.8634 - val_loss: 0.3753\n",
            "Epoch 9/10\n",
            "\u001b[1m1719/1719\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m8s\u001b[0m 4ms/step - accuracy: 0.8776 - loss: 0.3458 - val_accuracy: 0.8660 - val_loss: 0.3700\n",
            "Epoch 10/10\n",
            "\u001b[1m1719/1719\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m10s\u001b[0m 3ms/step - accuracy: 0.8804 - loss: 0.3361 - val_accuracy: 0.8674 - val_loss: 0.3650\n",
            "\u001b[1m313/313\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 2ms/step - accuracy: 0.8649 - loss: 0.3877\n",
            "Neurons: 300/300, Accuracy: 0.86080002784729\n",
            "Epoch 1/10\n",
            "\u001b[1m1719/1719\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m7s\u001b[0m 4ms/step - accuracy: 0.6717 - loss: 1.0416 - val_accuracy: 0.8240 - val_loss: 0.5064\n",
            "Epoch 2/10\n",
            "\u001b[1m1719/1719\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m10s\u001b[0m 4ms/step - accuracy: 0.8248 - loss: 0.5138 - val_accuracy: 0.8368 - val_loss: 0.4517\n",
            "Epoch 3/10\n",
            "\u001b[1m1719/1719\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m6s\u001b[0m 3ms/step - accuracy: 0.8422 - loss: 0.4581 - val_accuracy: 0.8456 - val_loss: 0.4287\n",
            "Epoch 4/10\n",
            "\u001b[1m1719/1719\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m12s\u001b[0m 4ms/step - accuracy: 0.8516 - loss: 0.4274 - val_accuracy: 0.8508 - val_loss: 0.4138\n",
            "Epoch 5/10\n",
            "\u001b[1m1719/1719\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m6s\u001b[0m 3ms/step - accuracy: 0.8586 - loss: 0.4055 - val_accuracy: 0.8530 - val_loss: 0.4022\n",
            "Epoch 6/10\n",
            "\u001b[1m1719/1719\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m7s\u001b[0m 4ms/step - accuracy: 0.8642 - loss: 0.3884 - val_accuracy: 0.8550 - val_loss: 0.3947\n",
            "Epoch 7/10\n",
            "\u001b[1m1719/1719\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m9s\u001b[0m 3ms/step - accuracy: 0.8680 - loss: 0.3741 - val_accuracy: 0.8580 - val_loss: 0.3852\n",
            "Epoch 8/10\n",
            "\u001b[1m1719/1719\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m6s\u001b[0m 4ms/step - accuracy: 0.8724 - loss: 0.3616 - val_accuracy: 0.8622 - val_loss: 0.3776\n",
            "Epoch 9/10\n",
            "\u001b[1m1719/1719\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m5s\u001b[0m 3ms/step - accuracy: 0.8751 - loss: 0.3505 - val_accuracy: 0.8638 - val_loss: 0.3708\n",
            "Epoch 10/10\n",
            "\u001b[1m1719/1719\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m6s\u001b[0m 4ms/step - accuracy: 0.8787 - loss: 0.3406 - val_accuracy: 0.8646 - val_loss: 0.3658\n",
            "\u001b[1m313/313\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 2ms/step - accuracy: 0.8629 - loss: 0.3922\n",
            "Neurons: 200/200, Accuracy: 0.8587999939918518\n",
            "Epoch 1/10\n",
            "\u001b[1m1719/1719\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m4s\u001b[0m 2ms/step - accuracy: 0.6469 - loss: 1.1273 - val_accuracy: 0.8200 - val_loss: 0.5329\n",
            "Epoch 2/10\n",
            "\u001b[1m1719/1719\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m6s\u001b[0m 2ms/step - accuracy: 0.8156 - loss: 0.5307 - val_accuracy: 0.8336 - val_loss: 0.4779\n",
            "Epoch 3/10\n",
            "\u001b[1m1719/1719\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m5s\u001b[0m 2ms/step - accuracy: 0.8351 - loss: 0.4742 - val_accuracy: 0.8408 - val_loss: 0.4503\n",
            "Epoch 4/10\n",
            "\u001b[1m1719/1719\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m5s\u001b[0m 2ms/step - accuracy: 0.8464 - loss: 0.4414 - val_accuracy: 0.8470 - val_loss: 0.4321\n",
            "Epoch 5/10\n",
            "\u001b[1m1719/1719\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m5s\u001b[0m 3ms/step - accuracy: 0.8541 - loss: 0.4188 - val_accuracy: 0.8490 - val_loss: 0.4176\n",
            "Epoch 6/10\n",
            "\u001b[1m1719/1719\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m4s\u001b[0m 2ms/step - accuracy: 0.8595 - loss: 0.4011 - val_accuracy: 0.8526 - val_loss: 0.4074\n",
            "Epoch 7/10\n",
            "\u001b[1m1719/1719\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m5s\u001b[0m 2ms/step - accuracy: 0.8633 - loss: 0.3869 - val_accuracy: 0.8556 - val_loss: 0.3992\n",
            "Epoch 8/10\n",
            "\u001b[1m1719/1719\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m6s\u001b[0m 3ms/step - accuracy: 0.8675 - loss: 0.3748 - val_accuracy: 0.8586 - val_loss: 0.3915\n",
            "Epoch 9/10\n",
            "\u001b[1m1719/1719\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m4s\u001b[0m 2ms/step - accuracy: 0.8708 - loss: 0.3644 - val_accuracy: 0.8626 - val_loss: 0.3852\n",
            "Epoch 10/10\n",
            "\u001b[1m1719/1719\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m4s\u001b[0m 2ms/step - accuracy: 0.8737 - loss: 0.3550 - val_accuracy: 0.8634 - val_loss: 0.3781\n",
            "\u001b[1m313/313\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 3ms/step - accuracy: 0.8584 - loss: 0.4037\n",
            "Neurons: 100/100, Accuracy: 0.8550000190734863\n",
            "Epoch 1/10\n",
            "\u001b[1m1719/1719\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m8s\u001b[0m 4ms/step - accuracy: 0.6866 - loss: 0.9807 - val_accuracy: 0.8344 - val_loss: 0.4995\n",
            "Epoch 2/10\n",
            "\u001b[1m1719/1719\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m7s\u001b[0m 4ms/step - accuracy: 0.8271 - loss: 0.5041 - val_accuracy: 0.8428 - val_loss: 0.4494\n",
            "Epoch 3/10\n",
            "\u001b[1m1719/1719\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m8s\u001b[0m 3ms/step - accuracy: 0.8444 - loss: 0.4516 - val_accuracy: 0.8504 - val_loss: 0.4262\n",
            "Epoch 4/10\n",
            "\u001b[1m1719/1719\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m7s\u001b[0m 4ms/step - accuracy: 0.8536 - loss: 0.4215 - val_accuracy: 0.8548 - val_loss: 0.4113\n",
            "Epoch 5/10\n",
            "\u001b[1m1719/1719\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m5s\u001b[0m 3ms/step - accuracy: 0.8607 - loss: 0.3998 - val_accuracy: 0.8574 - val_loss: 0.4013\n",
            "Epoch 6/10\n",
            "\u001b[1m1719/1719\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m10s\u001b[0m 3ms/step - accuracy: 0.8660 - loss: 0.3824 - val_accuracy: 0.8604 - val_loss: 0.3939\n",
            "Epoch 7/10\n",
            "\u001b[1m1719/1719\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m6s\u001b[0m 4ms/step - accuracy: 0.8695 - loss: 0.3680 - val_accuracy: 0.8618 - val_loss: 0.3882\n",
            "Epoch 8/10\n",
            "\u001b[1m1719/1719\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m9s\u001b[0m 3ms/step - accuracy: 0.8741 - loss: 0.3557 - val_accuracy: 0.8640 - val_loss: 0.3828\n",
            "Epoch 9/10\n",
            "\u001b[1m1719/1719\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m7s\u001b[0m 4ms/step - accuracy: 0.8782 - loss: 0.3448 - val_accuracy: 0.8658 - val_loss: 0.3763\n",
            "Epoch 10/10\n",
            "\u001b[1m1719/1719\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m10s\u001b[0m 4ms/step - accuracy: 0.8815 - loss: 0.3351 - val_accuracy: 0.8674 - val_loss: 0.3712\n",
            "\u001b[1m313/313\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 2ms/step - accuracy: 0.8619 - loss: 0.3910\n",
            "Neurons: 300/100, Accuracy: 0.8587999939918518\n",
            "Epoch 1/10\n",
            "\u001b[1m1719/1719\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m5s\u001b[0m 3ms/step - accuracy: 0.6465 - loss: 1.1430 - val_accuracy: 0.8210 - val_loss: 0.5313\n",
            "Epoch 2/10\n",
            "\u001b[1m1719/1719\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m6s\u001b[0m 3ms/step - accuracy: 0.8141 - loss: 0.5342 - val_accuracy: 0.8354 - val_loss: 0.4699\n",
            "Epoch 3/10\n",
            "\u001b[1m1719/1719\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m4s\u001b[0m 2ms/step - accuracy: 0.8351 - loss: 0.4749 - val_accuracy: 0.8424 - val_loss: 0.4421\n",
            "Epoch 4/10\n",
            "\u001b[1m1719/1719\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m6s\u001b[0m 3ms/step - accuracy: 0.8454 - loss: 0.4431 - val_accuracy: 0.8480 - val_loss: 0.4243\n",
            "Epoch 5/10\n",
            "\u001b[1m1719/1719\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m9s\u001b[0m 2ms/step - accuracy: 0.8536 - loss: 0.4204 - val_accuracy: 0.8520 - val_loss: 0.4122\n",
            "Epoch 6/10\n",
            "\u001b[1m1719/1719\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m7s\u001b[0m 3ms/step - accuracy: 0.8590 - loss: 0.4024 - val_accuracy: 0.8534 - val_loss: 0.4003\n",
            "Epoch 7/10\n",
            "\u001b[1m1719/1719\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m4s\u001b[0m 2ms/step - accuracy: 0.8640 - loss: 0.3877 - val_accuracy: 0.8582 - val_loss: 0.3936\n",
            "Epoch 8/10\n",
            "\u001b[1m1719/1719\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m4s\u001b[0m 2ms/step - accuracy: 0.8669 - loss: 0.3754 - val_accuracy: 0.8600 - val_loss: 0.3847\n",
            "Epoch 9/10\n",
            "\u001b[1m1719/1719\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m6s\u001b[0m 3ms/step - accuracy: 0.8710 - loss: 0.3645 - val_accuracy: 0.8598 - val_loss: 0.3784\n",
            "Epoch 10/10\n",
            "\u001b[1m1719/1719\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m4s\u001b[0m 2ms/step - accuracy: 0.8743 - loss: 0.3548 - val_accuracy: 0.8630 - val_loss: 0.3728\n",
            "\u001b[1m313/313\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 2ms/step - accuracy: 0.8574 - loss: 0.4007\n",
            "Neurons: 100/300, Accuracy: 0.8539000153541565\n"
          ]
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "###Exercise 3.\n",
        "\n",
        "Choose the best combination of inner layers and reset numberOfIterations to 30.\n",
        "\n",
        "Note what the accuracy is now."
      ],
      "metadata": {
        "id": "cQePcpRCJsEd"
      }
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "rvZhAKDCkc1m"
      },
      "source": [
        "### Using the model to make predictions"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "kUBy_xI_kc1m",
        "outputId": "ac434dcf-0ae2-4346-ee10-e0d50fe28320"
      },
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 57ms/step\n"
          ]
        },
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "array([[0.  , 0.  , 0.  , 0.  , 0.  , 0.26, 0.  , 0.12, 0.01, 0.6 ],\n",
              "       [0.  , 0.  , 0.95, 0.  , 0.01, 0.  , 0.04, 0.  , 0.  , 0.  ],\n",
              "       [0.  , 1.  , 0.  , 0.  , 0.  , 0.  , 0.  , 0.  , 0.  , 0.  ]],\n",
              "      dtype=float32)"
            ]
          },
          "metadata": {},
          "execution_count": 39
        }
      ],
      "source": [
        "X_new = X_test[:3]\n",
        "y_proba = model.predict(X_new)\n",
        "y_proba.round(2)"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "LULTGTCckc1m",
        "outputId": "d0284e30-b4eb-4c6f-baa7-bf650e2fc7c9"
      },
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "array([9, 2, 1])"
            ]
          },
          "metadata": {},
          "execution_count": 40
        }
      ],
      "source": [
        "y_pred = y_proba.argmax(axis=-1)\n",
        "y_pred"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "DyBJYXzgkc1n",
        "outputId": "7b88f3fd-c7b7-4690-e458-f5379ce6b224"
      },
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "array(['Ankle boot', 'Pullover', 'Trouser'], dtype='<U11')"
            ]
          },
          "metadata": {},
          "execution_count": 41
        }
      ],
      "source": [
        "np.array(class_names)[y_pred]"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "8vnFqg5akc1n",
        "outputId": "6a77f8ea-babe-4655-8a3e-4e50075bbc2d"
      },
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "array([9, 2, 1], dtype=uint8)"
            ]
          },
          "metadata": {},
          "execution_count": 42
        }
      ],
      "source": [
        "y_new = y_test[:3]\n",
        "y_new"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 224
        },
        "id": "REG3lO92kc1n",
        "outputId": "aea0877a-5be0-45b5-d0ed-70a1cf09c0c4"
      },
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 720x240 with 3 Axes>"
            ],
            "image/png": "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\n"
          },
          "metadata": {}
        }
      ],
      "source": [
        "# extra code – this cell generates and saves Figure 10–12\n",
        "plt.figure(figsize=(7.2, 2.4))\n",
        "for index, image in enumerate(X_new):\n",
        "    plt.subplot(1, 3, index + 1)\n",
        "    plt.imshow(image, cmap=\"binary\", interpolation=\"nearest\")\n",
        "    plt.axis('off')\n",
        "    plt.title(class_names[y_test[index]])\n",
        "plt.subplots_adjust(wspace=0.2, hspace=0.5)\n",
        "# save_fig('fashion_mnist_images_plot', tight_layout=False)\n",
        "plt.show()"
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "###Exercise 4"
      ],
      "metadata": {
        "id": "tQIiOrq9Mtjt"
      }
    },
    {
      "cell_type": "markdown",
      "source": [
        "1. What is the idea behind MLP ?\n",
        "\n",
        "MLP, eller Multilayer Perceptron, er en feedforward-neuralt netværk, der skaæ bestå af et eller flere lag af neuroner. Forståelsen af koncepten er at bruge flere lag af kunstige neuroner til at lære og forstå komplekse funktioner ved at opdage mønstre i data.\n",
        "\n",
        "\n",
        "MLP skal forståes som det fungerer ved at tage inddata, behandle dem gennem flere lag af neuroner, der hver især udfører beregninger, og derefter levere et output. Hvert lag hjælper med at opfange mere komplekse mønstre i dataene, og netværket lærer gradvist at forbedre sine forudsigelser gennem gentagne tilpasninger."
      ],
      "metadata": {
        "id": "88J5CezxM1T-"
      }
    },
    {
      "cell_type": "markdown",
      "source": [
        "2. What is backpropagation ?\n",
        "\n",
        "Backpropagation er en metode inden for at træne af kunstige neurale netværk, det bruges til at justere vægtene i netværket med henblik på at minimere fejl i modelens forudsigelser."
      ],
      "metadata": {
        "id": "8NMhCZQhNVBP"
      }
    },
    {
      "cell_type": "markdown",
      "source": [
        "3. What is an activation function the purpose of it and and which types do you know ?\n",
        "\n",
        "\n",
        "En aktiveringsfunktion i et kunstigt neuralt netværk er en matematisk funktion, der anvendes på outputtet af hver neuron for at introducere ikke-linearitet. Aktiveringsfunktionen bestemmer, om en neuron \"aktiveres\" og dermed sender sit signal videre til de næste lag i netværket.\n",
        "\n",
        "\n",
        "\n",
        "Sigmoid-funktion:\n",
        "\n",
        "Output: Mellem 0 og 1.\n",
        "Brug: Tidligere meget brugt i klassificeringsopgaver, men har mistet popularitet pga. problemet med vanishing gradients, hvor gradienterne bliver meget små, hvilket gør det svært at træne dybe netværk.\n",
        "Fordel: Output kan tolkes som en sandsynlighed.\n",
        "\n",
        "Tanh (hyperbolsk tangent):\n",
        "\n",
        "Output: Mellem -1 og 1.\n",
        "Brug: Ligner sigmoid, men output er centreret omkring 0, hvilket gør det mere velegnet i praksis end sigmoid i visse tilfælde.\n",
        "Fordel: Tanh kan håndtere negative input bedre end sigmoid.\n",
        "\n",
        "\n",
        "ReLU (Rectified Linear Unit):\n",
        "\n",
        "Brug: Meget populær i moderne neurale netværk, især dybe netværk.\n",
        "Fordel: Let at beregne og hjælper med at forhindre problemet med vanishing gradients.\n",
        "Ulempe: Kan føre til \"døde\" neuroner, hvor nogle neuroner aldrig aktiveres.\n",
        "\n",
        "Leaky ReLU:\n",
        "Formel:\n",
        "Output: Tillader en lille, ikke-nul gradient, når\n",
        "x≤0\n",
        "x≤0.\n",
        "Brug: En forbedring af ReLU, der hjælper med at forhindre \"døde\" neuroner.\n",
        "\n",
        "Softmax:\n",
        "  for hver kategori\n",
        "i\n",
        "i.\n",
        "Output: En sandsynlighedsfordeling (summen af outputtene er 1).\n",
        "Brug: Bruges typisk i outputlaget af klassifikationsmodeller, især i multi-class klassificering.\n",
        "Fordel: Giver et sandsynlighedsoutput, der kan bruges til at afgøre den mest sandsynlige klasse.\n",
        "\n",
        "\n",
        "Swish:\n",
        "Brug: En nyere aktiveringsfunktion, der er foreslået af Google og viser bedre resultater end ReLU i nogle opgaver.\n",
        "Fordel: Glattere end ReLU og kan resultere i bedre læring i visse situationer."
      ],
      "metadata": {
        "id": "LM_NnZL2NfdJ"
      }
    }
  ],
  "metadata": {
    "colab": {
      "provenance": [],
      "include_colab_link": true
    },
    "kernelspec": {
      "display_name": "Python 3",
      "language": "python",
      "name": "python3"
    },
    "language_info": {
      "codemirror_mode": {
        "name": "ipython",
        "version": 3
      },
      "file_extension": ".py",
      "mimetype": "text/x-python",
      "name": "python",
      "nbconvert_exporter": "python",
      "pygments_lexer": "ipython3",
      "version": "3.10.6"
    },
    "nav_menu": {
      "height": "264px",
      "width": "369px"
    },
    "toc": {
      "navigate_menu": true,
      "number_sections": true,
      "sideBar": true,
      "threshold": 6,
      "toc_cell": false,
      "toc_section_display": "block",
      "toc_window_display": false
    }
  },
  "nbformat": 4,
  "nbformat_minor": 0
}
No results found