02_neural_network_classification_exercises.ipynb

{
  "nbformat": 4,
  "nbformat_minor": 0,
  "metadata": {
    "colab": {
      "provenance": [],
      "collapsed_sections": [],
      "toc_visible": true,
      "authorship_tag": "ABX9TyNLjJlAfFB3wW0dFq0RSroL",
      "include_colab_link": true
    },
    "kernelspec": {
      "name": "python3",
      "display_name": "Python 3"
    },
    "language_info": {
      "name": "python"
    }
  },
  "cells": [
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "view-in-github",
        "colab_type": "text"
      },
      "source": [
        "<a href=\"https://colab.research.google.com/github/SaketMunda/tensorflow-fundamentals/blob/master/02_neural_network_classification_exercises.ipynb\" target=\"_parent\"><img src=\"https://colab.research.google.com/assets/colab-badge.svg\" alt=\"Open In Colab\"/></a>"
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "# Neural Network Classification with TensorFlow Exercises\n",
        "\n",
        "This notebook consist of various exercises that are listed in the [readMe](https://github.com/SaketMunda/tensorflow-fundamentals/blob/master/README.md) of the [Github Repo](https://github.com/SaketMunda/tensorflow-fundamentals)"
      ],
      "metadata": {
        "id": "VLXCyCmB-YBg"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "# Import required libraries\n",
        "import tensorflow as tf\n",
        "import numpy as np\n",
        "import pandas as pd\n",
        "import matplotlib.pyplot as plt"
      ],
      "metadata": {
        "id": "ul51uC4M_smZ"
      },
      "execution_count": 1,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "source": [
        "## 2. Replicate the model pictured in the TensorFlow Playground diagram \n",
        "\n",
        "Given below is the diagram, use TensorFlow code and Compile it using the Adam optimizer, binary crossentropy loss and accuracy metric. Once it's compiled check a summary of the model.\n",
        "\n",
        "![image](https://raw.githubusercontent.com/mrdbourke/tensorflow-deep-learning/main/images/02-tensorflow-playground-replication-exercise.png)"
      ],
      "metadata": {
        "id": "TFm3xq1D-18n"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "# First need to create some dataset for this, X (data) and y (labels)\n",
        "from sklearn.datasets import make_circles\n",
        "\n",
        "# Make 1000 samples\n",
        "n_samples = 1000\n",
        "X, y = make_circles(n_samples, noise= 0.03, random_state = 17)"
      ],
      "metadata": {
        "id": "w_jKznEa_NWP"
      },
      "execution_count": 2,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "# Let's check the samples\n",
        "X[:5], y[:5]"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "2_b_yXd4Acz2",
        "outputId": "f88710be-8b5a-4818-bb45-3ab5299e9624"
      },
      "execution_count": 3,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "(array([[-0.69661282, -0.32845725],\n",
              "        [ 0.95581012, -0.05841444],\n",
              "        [ 0.53469756, -0.57338982],\n",
              "        [-0.89332834,  0.42098993],\n",
              "        [-0.65647965, -0.4353153 ]]), array([1, 0, 1, 0, 1]))"
            ]
          },
          "metadata": {},
          "execution_count": 3
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "# Shape of the samples\n",
        "X.shape, y.shape"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "fsqxHBoDAh-K",
        "outputId": "05b831ba-487d-484a-9b43-079a0cbadbce"
      },
      "execution_count": 4,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "((1000, 2), (1000,))"
            ]
          },
          "metadata": {},
          "execution_count": 4
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "Okay what we learned from the data ? Seeing the data in this format is not speaking out anything. Let's visualize it."
      ],
      "metadata": {
        "id": "IlWp6FDYAmne"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "circles = pd.DataFrame({\"X0\":X[:, 0], \"X1\": X[:, 1], \"label\": y})\n",
        "circles.head(10)"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 363
        },
        "id": "LhE1aTsSArdm",
        "outputId": "9ed88cd1-543a-40b0-e3bd-7262d2e7c28a"
      },
      "execution_count": 5,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "         X0        X1  label\n",
              "0 -0.696613 -0.328457      1\n",
              "1  0.955810 -0.058414      0\n",
              "2  0.534698 -0.573390      1\n",
              "3 -0.893328  0.420990      0\n",
              "4 -0.656480 -0.435315      1\n",
              "5  0.611903  0.800231      0\n",
              "6 -0.598358 -0.567453      1\n",
              "7  0.331679  0.922320      0\n",
              "8  0.782751 -0.135743      1\n",
              "9  0.579686  0.842586      0"
            ],
            "text/html": [
              "\n",
              "  <div id=\"df-cbbe0ae2-811b-4310-a1f9-a6d5ee7fad90\">\n",
              "    <div class=\"colab-df-container\">\n",
              "      <div>\n",
              "<style scoped>\n",
              "    .dataframe tbody tr th:only-of-type {\n",
              "        vertical-align: middle;\n",
              "    }\n",
              "\n",
              "    .dataframe tbody tr th {\n",
              "        vertical-align: top;\n",
              "    }\n",
              "\n",
              "    .dataframe thead th {\n",
              "        text-align: right;\n",
              "    }\n",
              "</style>\n",
              "<table border=\"1\" class=\"dataframe\">\n",
              "  <thead>\n",
              "    <tr style=\"text-align: right;\">\n",
              "      <th></th>\n",
              "      <th>X0</th>\n",
              "      <th>X1</th>\n",
              "      <th>label</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>0</th>\n",
              "      <td>-0.696613</td>\n",
              "      <td>-0.328457</td>\n",
              "      <td>1</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1</th>\n",
              "      <td>0.955810</td>\n",
              "      <td>-0.058414</td>\n",
              "      <td>0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2</th>\n",
              "      <td>0.534698</td>\n",
              "      <td>-0.573390</td>\n",
              "      <td>1</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>3</th>\n",
              "      <td>-0.893328</td>\n",
              "      <td>0.420990</td>\n",
              "      <td>0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>4</th>\n",
              "      <td>-0.656480</td>\n",
              "      <td>-0.435315</td>\n",
              "      <td>1</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>5</th>\n",
              "      <td>0.611903</td>\n",
              "      <td>0.800231</td>\n",
              "      <td>0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>6</th>\n",
              "      <td>-0.598358</td>\n",
              "      <td>-0.567453</td>\n",
              "      <td>1</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>7</th>\n",
              "      <td>0.331679</td>\n",
              "      <td>0.922320</td>\n",
              "      <td>0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>8</th>\n",
              "      <td>0.782751</td>\n",
              "      <td>-0.135743</td>\n",
              "      <td>1</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>9</th>\n",
              "      <td>0.579686</td>\n",
              "      <td>0.842586</td>\n",
              "      <td>0</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>\n",
              "      <button class=\"colab-df-convert\" onclick=\"convertToInteractive('df-cbbe0ae2-811b-4310-a1f9-a6d5ee7fad90')\"\n",
              "              title=\"Convert this dataframe to an interactive table.\"\n",
              "              style=\"display:none;\">\n",
              "        \n",
              "  <svg xmlns=\"http://www.w3.org/2000/svg\" height=\"24px\"viewBox=\"0 0 24 24\"\n",
              "       width=\"24px\">\n",
              "    <path d=\"M0 0h24v24H0V0z\" fill=\"none\"/>\n",
              "    <path d=\"M18.56 5.44l.94 2.06.94-2.06 2.06-.94-2.06-.94-.94-2.06-.94 2.06-2.06.94zm-11 1L8.5 8.5l.94-2.06 2.06-.94-2.06-.94L8.5 2.5l-.94 2.06-2.06.94zm10 10l.94 2.06.94-2.06 2.06-.94-2.06-.94-.94-2.06-.94 2.06-2.06.94z\"/><path d=\"M17.41 7.96l-1.37-1.37c-.4-.4-.92-.59-1.43-.59-.52 0-1.04.2-1.43.59L10.3 9.45l-7.72 7.72c-.78.78-.78 2.05 0 2.83L4 21.41c.39.39.9.59 1.41.59.51 0 1.02-.2 1.41-.59l7.78-7.78 2.81-2.81c.8-.78.8-2.07 0-2.86zM5.41 20L4 18.59l7.72-7.72 1.47 1.35L5.41 20z\"/>\n",
              "  </svg>\n",
              "      </button>\n",
              "      \n",
              "  <style>\n",
              "    .colab-df-container {\n",
              "      display:flex;\n",
              "      flex-wrap:wrap;\n",
              "      gap: 12px;\n",
              "    }\n",
              "\n",
              "    .colab-df-convert {\n",
              "      background-color: #E8F0FE;\n",
              "      border: none;\n",
              "      border-radius: 50%;\n",
              "      cursor: pointer;\n",
              "      display: none;\n",
              "      fill: #1967D2;\n",
              "      height: 32px;\n",
              "      padding: 0 0 0 0;\n",
              "      width: 32px;\n",
              "    }\n",
              "\n",
              "    .colab-df-convert:hover {\n",
              "      background-color: #E2EBFA;\n",
              "      box-shadow: 0px 1px 2px rgba(60, 64, 67, 0.3), 0px 1px 3px 1px rgba(60, 64, 67, 0.15);\n",
              "      fill: #174EA6;\n",
              "    }\n",
              "\n",
              "    [theme=dark] .colab-df-convert {\n",
              "      background-color: #3B4455;\n",
              "      fill: #D2E3FC;\n",
              "    }\n",
              "\n",
              "    [theme=dark] .colab-df-convert:hover {\n",
              "      background-color: #434B5C;\n",
              "      box-shadow: 0px 1px 3px 1px rgba(0, 0, 0, 0.15);\n",
              "      filter: drop-shadow(0px 1px 2px rgba(0, 0, 0, 0.3));\n",
              "      fill: #FFFFFF;\n",
              "    }\n",
              "  </style>\n",
              "\n",
              "      <script>\n",
              "        const buttonEl =\n",
              "          document.querySelector('#df-cbbe0ae2-811b-4310-a1f9-a6d5ee7fad90 button.colab-df-convert');\n",
              "        buttonEl.style.display =\n",
              "          google.colab.kernel.accessAllowed ? 'block' : 'none';\n",
              "\n",
              "        async function convertToInteractive(key) {\n",
              "          const element = document.querySelector('#df-cbbe0ae2-811b-4310-a1f9-a6d5ee7fad90');\n",
              "          const dataTable =\n",
              "            await google.colab.kernel.invokeFunction('convertToInteractive',\n",
              "                                                     [key], {});\n",
              "          if (!dataTable) return;\n",
              "\n",
              "          const docLinkHtml = 'Like what you see? Visit the ' +\n",
              "            '<a target=\"_blank\" href=https://colab.research.google.com/notebooks/data_table.ipynb>data table notebook</a>'\n",
              "            + ' to learn more about interactive tables.';\n",
              "          element.innerHTML = '';\n",
              "          dataTable['output_type'] = 'display_data';\n",
              "          await google.colab.output.renderOutput(dataTable, element);\n",
              "          const docLink = document.createElement('div');\n",
              "          docLink.innerHTML = docLinkHtml;\n",
              "          element.appendChild(docLink);\n",
              "        }\n",
              "      </script>\n",
              "    </div>\n",
              "  </div>\n",
              "  "
            ]
          },
          "metadata": {},
          "execution_count": 5
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "# If we use this dataframe to plot\n",
        "plt.scatter(circles[\"X0\"], circles[\"X1\"], c=circles[\"label\"]);"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 265
        },
        "id": "mzLNxsc0A1T-",
        "outputId": "487ff40b-825f-4c91-f9bb-d6f2f818fcd9"
      },
      "execution_count": 6,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 432x288 with 1 Axes>"
            ],
            "image/png": "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\n"
          },
          "metadata": {
            "needs_background": "light"
          }
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "# check out the different labels\n",
        "circles.label.value_counts()"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "KxQeXH0FBvpH",
        "outputId": "24c5a7ea-6784-4628-e149-7687baf14263"
      },
      "execution_count": 7,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "1    500\n",
              "0    500\n",
              "Name: label, dtype: int64"
            ]
          },
          "metadata": {},
          "execution_count": 7
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "We can see that we have label `0` and `1`, it means that we're going to solve a binary classification problem."
      ],
      "metadata": {
        "id": "YMPaI3tqB9mP"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "# Let's split our data into training and test set, ratio will be 80-20\n",
        "from sklearn.model_selection import train_test_split\n",
        "\n",
        "X_train, X_test, y_train, y_test = train_test_split(X,y,test_size=0.2)"
      ],
      "metadata": {
        "id": "aqXPG2-QCFxM"
      },
      "execution_count": 8,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "# Check the shapes of training and test samples\n",
        "X_train.shape, X_test.shape, y_train.shape, y_test.shape"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "YN4cIzHiCrqA",
        "outputId": "3d0abf3a-8b55-4cf1-b2e6-839f37aa7b9a"
      },
      "execution_count": 9,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "((800, 2), (200, 2), (800,), (200,))"
            ]
          },
          "metadata": {},
          "execution_count": 9
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "Alright, now we are ready to build our model and replicate the layers shown above in the TensorFlow playground screenshot."
      ],
      "metadata": {
        "id": "Vji-nuhGDDMD"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "# set the random seed\n",
        "tf.random.set_seed(17)\n",
        "\n",
        "# Create a model with 5 hidden layers having 6 neuron each layer with activation as ReLu and 1 output layer with activation as sigmoid\n",
        "model_1 = tf.keras.Sequential([\n",
        "    tf.keras.layers.Dense(6, activation='relu'),\n",
        "    tf.keras.layers.Dense(6, activation='relu'),\n",
        "    tf.keras.layers.Dense(6, activation='relu'),\n",
        "    tf.keras.layers.Dense(6, activation='relu'),\n",
        "    tf.keras.layers.Dense(6, activation='relu'),\n",
        "    tf.keras.layers.Dense(1, activation='sigmoid')\n",
        "])\n",
        "\n",
        "# Compile a model\n",
        "model_1.compile(loss='binary_crossentropy',\n",
        "                optimizer='Adam',\n",
        "                metrics=['accuracy'])\n",
        "\n",
        "# Fit the model\n",
        "history_1 = model_1.fit(X_train,\n",
        "                        y_train,\n",
        "                        epochs=100,\n",
        "                        validation_data=(X_test, y_test))"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "ppI0yri9DOqj",
        "outputId": "a227cd74-8d69-4435-cbea-77f320f706d7"
      },
      "execution_count": 10,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Epoch 1/100\n",
            "25/25 [==============================] - 1s 10ms/step - loss: 0.6906 - accuracy: 0.4963 - val_loss: 0.6904 - val_accuracy: 0.4950\n",
            "Epoch 2/100\n",
            "25/25 [==============================] - 0s 2ms/step - loss: 0.6896 - accuracy: 0.5163 - val_loss: 0.6902 - val_accuracy: 0.5050\n",
            "Epoch 3/100\n",
            "25/25 [==============================] - 0s 2ms/step - loss: 0.6890 - accuracy: 0.5188 - val_loss: 0.6900 - val_accuracy: 0.5100\n",
            "Epoch 4/100\n",
            "25/25 [==============================] - 0s 3ms/step - loss: 0.6883 - accuracy: 0.5175 - val_loss: 0.6896 - val_accuracy: 0.5100\n",
            "Epoch 5/100\n",
            "25/25 [==============================] - 0s 3ms/step - loss: 0.6875 - accuracy: 0.5200 - val_loss: 0.6893 - val_accuracy: 0.5050\n",
            "Epoch 6/100\n",
            "25/25 [==============================] - 0s 2ms/step - loss: 0.6867 - accuracy: 0.5375 - val_loss: 0.6885 - val_accuracy: 0.5050\n",
            "Epoch 7/100\n",
            "25/25 [==============================] - 0s 3ms/step - loss: 0.6855 - accuracy: 0.5362 - val_loss: 0.6880 - val_accuracy: 0.5100\n",
            "Epoch 8/100\n",
            "25/25 [==============================] - 0s 3ms/step - loss: 0.6840 - accuracy: 0.5387 - val_loss: 0.6871 - val_accuracy: 0.4900\n",
            "Epoch 9/100\n",
            "25/25 [==============================] - 0s 2ms/step - loss: 0.6824 - accuracy: 0.5663 - val_loss: 0.6863 - val_accuracy: 0.5050\n",
            "Epoch 10/100\n",
            "25/25 [==============================] - 0s 2ms/step - loss: 0.6801 - accuracy: 0.5600 - val_loss: 0.6848 - val_accuracy: 0.4950\n",
            "Epoch 11/100\n",
            "25/25 [==============================] - 0s 3ms/step - loss: 0.6778 - accuracy: 0.5650 - val_loss: 0.6834 - val_accuracy: 0.5000\n",
            "Epoch 12/100\n",
            "25/25 [==============================] - 0s 2ms/step - loss: 0.6748 - accuracy: 0.5738 - val_loss: 0.6805 - val_accuracy: 0.4850\n",
            "Epoch 13/100\n",
            "25/25 [==============================] - 0s 2ms/step - loss: 0.6712 - accuracy: 0.5575 - val_loss: 0.6763 - val_accuracy: 0.5200\n",
            "Epoch 14/100\n",
            "25/25 [==============================] - 0s 2ms/step - loss: 0.6672 - accuracy: 0.5612 - val_loss: 0.6715 - val_accuracy: 0.5500\n",
            "Epoch 15/100\n",
            "25/25 [==============================] - 0s 2ms/step - loss: 0.6600 - accuracy: 0.6150 - val_loss: 0.6645 - val_accuracy: 0.5350\n",
            "Epoch 16/100\n",
            "25/25 [==============================] - 0s 2ms/step - loss: 0.6528 - accuracy: 0.6137 - val_loss: 0.6574 - val_accuracy: 0.5400\n",
            "Epoch 17/100\n",
            "25/25 [==============================] - 0s 2ms/step - loss: 0.6433 - accuracy: 0.6687 - val_loss: 0.6454 - val_accuracy: 0.5650\n",
            "Epoch 18/100\n",
            "25/25 [==============================] - 0s 3ms/step - loss: 0.6302 - accuracy: 0.6862 - val_loss: 0.6386 - val_accuracy: 0.5200\n",
            "Epoch 19/100\n",
            "25/25 [==============================] - 0s 2ms/step - loss: 0.6150 - accuracy: 0.7025 - val_loss: 0.6169 - val_accuracy: 0.6650\n",
            "Epoch 20/100\n",
            "25/25 [==============================] - 0s 2ms/step - loss: 0.5975 - accuracy: 0.7287 - val_loss: 0.6066 - val_accuracy: 0.6200\n",
            "Epoch 21/100\n",
            "25/25 [==============================] - 0s 2ms/step - loss: 0.5812 - accuracy: 0.7312 - val_loss: 0.5843 - val_accuracy: 0.7100\n",
            "Epoch 22/100\n",
            "25/25 [==============================] - 0s 2ms/step - loss: 0.5635 - accuracy: 0.7450 - val_loss: 0.5712 - val_accuracy: 0.7450\n",
            "Epoch 23/100\n",
            "25/25 [==============================] - 0s 3ms/step - loss: 0.5491 - accuracy: 0.7487 - val_loss: 0.5534 - val_accuracy: 0.7350\n",
            "Epoch 24/100\n",
            "25/25 [==============================] - 0s 2ms/step - loss: 0.5307 - accuracy: 0.7663 - val_loss: 0.5485 - val_accuracy: 0.7100\n",
            "Epoch 25/100\n",
            "25/25 [==============================] - 0s 2ms/step - loss: 0.5166 - accuracy: 0.7663 - val_loss: 0.5247 - val_accuracy: 0.7400\n",
            "Epoch 26/100\n",
            "25/25 [==============================] - 0s 2ms/step - loss: 0.4984 - accuracy: 0.7788 - val_loss: 0.5171 - val_accuracy: 0.7600\n",
            "Epoch 27/100\n",
            "25/25 [==============================] - 0s 3ms/step - loss: 0.4871 - accuracy: 0.7837 - val_loss: 0.5022 - val_accuracy: 0.7400\n",
            "Epoch 28/100\n",
            "25/25 [==============================] - 0s 3ms/step - loss: 0.4742 - accuracy: 0.7788 - val_loss: 0.4955 - val_accuracy: 0.7450\n",
            "Epoch 29/100\n",
            "25/25 [==============================] - 0s 3ms/step - loss: 0.4585 - accuracy: 0.8037 - val_loss: 0.4832 - val_accuracy: 0.7550\n",
            "Epoch 30/100\n",
            "25/25 [==============================] - 0s 3ms/step - loss: 0.4505 - accuracy: 0.8075 - val_loss: 0.4765 - val_accuracy: 0.7700\n",
            "Epoch 31/100\n",
            "25/25 [==============================] - 0s 3ms/step - loss: 0.4394 - accuracy: 0.8125 - val_loss: 0.4693 - val_accuracy: 0.7600\n",
            "Epoch 32/100\n",
            "25/25 [==============================] - 0s 2ms/step - loss: 0.4302 - accuracy: 0.8138 - val_loss: 0.4606 - val_accuracy: 0.7800\n",
            "Epoch 33/100\n",
            "25/25 [==============================] - 0s 2ms/step - loss: 0.4235 - accuracy: 0.8150 - val_loss: 0.4578 - val_accuracy: 0.7800\n",
            "Epoch 34/100\n",
            "25/25 [==============================] - 0s 2ms/step - loss: 0.4135 - accuracy: 0.8200 - val_loss: 0.4470 - val_accuracy: 0.7800\n",
            "Epoch 35/100\n",
            "25/25 [==============================] - 0s 2ms/step - loss: 0.4055 - accuracy: 0.8250 - val_loss: 0.4414 - val_accuracy: 0.7800\n",
            "Epoch 36/100\n",
            "25/25 [==============================] - 0s 3ms/step - loss: 0.3992 - accuracy: 0.8288 - val_loss: 0.4336 - val_accuracy: 0.7900\n",
            "Epoch 37/100\n",
            "25/25 [==============================] - 0s 3ms/step - loss: 0.3936 - accuracy: 0.8275 - val_loss: 0.4268 - val_accuracy: 0.8000\n",
            "Epoch 38/100\n",
            "25/25 [==============================] - 0s 2ms/step - loss: 0.3856 - accuracy: 0.8313 - val_loss: 0.4207 - val_accuracy: 0.7850\n",
            "Epoch 39/100\n",
            "25/25 [==============================] - 0s 2ms/step - loss: 0.3778 - accuracy: 0.8363 - val_loss: 0.4113 - val_accuracy: 0.7900\n",
            "Epoch 40/100\n",
            "25/25 [==============================] - 0s 2ms/step - loss: 0.3709 - accuracy: 0.8388 - val_loss: 0.4021 - val_accuracy: 0.8100\n",
            "Epoch 41/100\n",
            "25/25 [==============================] - 0s 3ms/step - loss: 0.3662 - accuracy: 0.8438 - val_loss: 0.3973 - val_accuracy: 0.8100\n",
            "Epoch 42/100\n",
            "25/25 [==============================] - 0s 2ms/step - loss: 0.3604 - accuracy: 0.8450 - val_loss: 0.3984 - val_accuracy: 0.8000\n",
            "Epoch 43/100\n",
            "25/25 [==============================] - 0s 2ms/step - loss: 0.3552 - accuracy: 0.8462 - val_loss: 0.3899 - val_accuracy: 0.8100\n",
            "Epoch 44/100\n",
            "25/25 [==============================] - 0s 2ms/step - loss: 0.3515 - accuracy: 0.8475 - val_loss: 0.3961 - val_accuracy: 0.8000\n",
            "Epoch 45/100\n",
            "25/25 [==============================] - 0s 2ms/step - loss: 0.3424 - accuracy: 0.8487 - val_loss: 0.3757 - val_accuracy: 0.8200\n",
            "Epoch 46/100\n",
            "25/25 [==============================] - 0s 2ms/step - loss: 0.3375 - accuracy: 0.8487 - val_loss: 0.3690 - val_accuracy: 0.8200\n",
            "Epoch 47/100\n",
            "25/25 [==============================] - 0s 2ms/step - loss: 0.3307 - accuracy: 0.8537 - val_loss: 0.3644 - val_accuracy: 0.8250\n",
            "Epoch 48/100\n",
            "25/25 [==============================] - 0s 2ms/step - loss: 0.3268 - accuracy: 0.8525 - val_loss: 0.3623 - val_accuracy: 0.8150\n",
            "Epoch 49/100\n",
            "25/25 [==============================] - 0s 3ms/step - loss: 0.3198 - accuracy: 0.8562 - val_loss: 0.3589 - val_accuracy: 0.8150\n",
            "Epoch 50/100\n",
            "25/25 [==============================] - 0s 2ms/step - loss: 0.3172 - accuracy: 0.8562 - val_loss: 0.3548 - val_accuracy: 0.8100\n",
            "Epoch 51/100\n",
            "25/25 [==============================] - 0s 2ms/step - loss: 0.3063 - accuracy: 0.8600 - val_loss: 0.3487 - val_accuracy: 0.8150\n",
            "Epoch 52/100\n",
            "25/25 [==============================] - 0s 2ms/step - loss: 0.3036 - accuracy: 0.8600 - val_loss: 0.3388 - val_accuracy: 0.8250\n",
            "Epoch 53/100\n",
            "25/25 [==============================] - 0s 2ms/step - loss: 0.2974 - accuracy: 0.8612 - val_loss: 0.3246 - val_accuracy: 0.8350\n",
            "Epoch 54/100\n",
            "25/25 [==============================] - 0s 2ms/step - loss: 0.2893 - accuracy: 0.8637 - val_loss: 0.3193 - val_accuracy: 0.8350\n",
            "Epoch 55/100\n",
            "25/25 [==============================] - 0s 3ms/step - loss: 0.2837 - accuracy: 0.8637 - val_loss: 0.3170 - val_accuracy: 0.8300\n",
            "Epoch 56/100\n",
            "25/25 [==============================] - 0s 3ms/step - loss: 0.2790 - accuracy: 0.8662 - val_loss: 0.3048 - val_accuracy: 0.8400\n",
            "Epoch 57/100\n",
            "25/25 [==============================] - 0s 3ms/step - loss: 0.2743 - accuracy: 0.8650 - val_loss: 0.2975 - val_accuracy: 0.8400\n",
            "Epoch 58/100\n",
            "25/25 [==============================] - 0s 3ms/step - loss: 0.2678 - accuracy: 0.8687 - val_loss: 0.2915 - val_accuracy: 0.8400\n",
            "Epoch 59/100\n",
            "25/25 [==============================] - 0s 2ms/step - loss: 0.2625 - accuracy: 0.8700 - val_loss: 0.2818 - val_accuracy: 0.8450\n",
            "Epoch 60/100\n",
            "25/25 [==============================] - 0s 2ms/step - loss: 0.2589 - accuracy: 0.8687 - val_loss: 0.2710 - val_accuracy: 0.8550\n",
            "Epoch 61/100\n",
            "25/25 [==============================] - 0s 3ms/step - loss: 0.2549 - accuracy: 0.8687 - val_loss: 0.2730 - val_accuracy: 0.8450\n",
            "Epoch 62/100\n",
            "25/25 [==============================] - 0s 3ms/step - loss: 0.2541 - accuracy: 0.8725 - val_loss: 0.2643 - val_accuracy: 0.8500\n",
            "Epoch 63/100\n",
            "25/25 [==============================] - 0s 2ms/step - loss: 0.2463 - accuracy: 0.8725 - val_loss: 0.2561 - val_accuracy: 0.8550\n",
            "Epoch 64/100\n",
            "25/25 [==============================] - 0s 2ms/step - loss: 0.2434 - accuracy: 0.8737 - val_loss: 0.2568 - val_accuracy: 0.8500\n",
            "Epoch 65/100\n",
            "25/25 [==============================] - 0s 2ms/step - loss: 0.2401 - accuracy: 0.8712 - val_loss: 0.2480 - val_accuracy: 0.8600\n",
            "Epoch 66/100\n",
            "25/25 [==============================] - 0s 2ms/step - loss: 0.2337 - accuracy: 0.8788 - val_loss: 0.2534 - val_accuracy: 0.8550\n",
            "Epoch 67/100\n",
            "25/25 [==============================] - 0s 2ms/step - loss: 0.2297 - accuracy: 0.8775 - val_loss: 0.2437 - val_accuracy: 0.8600\n",
            "Epoch 68/100\n",
            "25/25 [==============================] - 0s 3ms/step - loss: 0.2265 - accuracy: 0.8813 - val_loss: 0.2440 - val_accuracy: 0.8600\n",
            "Epoch 69/100\n",
            "25/25 [==============================] - 0s 2ms/step - loss: 0.2233 - accuracy: 0.8813 - val_loss: 0.2413 - val_accuracy: 0.8550\n",
            "Epoch 70/100\n",
            "25/25 [==============================] - 0s 3ms/step - loss: 0.2204 - accuracy: 0.8813 - val_loss: 0.2346 - val_accuracy: 0.8600\n",
            "Epoch 71/100\n",
            "25/25 [==============================] - 0s 3ms/step - loss: 0.2189 - accuracy: 0.8800 - val_loss: 0.2348 - val_accuracy: 0.8550\n",
            "Epoch 72/100\n",
            "25/25 [==============================] - 0s 2ms/step - loss: 0.2153 - accuracy: 0.8825 - val_loss: 0.2311 - val_accuracy: 0.8600\n",
            "Epoch 73/100\n",
            "25/25 [==============================] - 0s 3ms/step - loss: 0.2151 - accuracy: 0.8825 - val_loss: 0.2273 - val_accuracy: 0.8650\n",
            "Epoch 74/100\n",
            "25/25 [==============================] - 0s 2ms/step - loss: 0.2123 - accuracy: 0.8825 - val_loss: 0.2293 - val_accuracy: 0.8650\n",
            "Epoch 75/100\n",
            "25/25 [==============================] - 0s 2ms/step - loss: 0.2091 - accuracy: 0.8838 - val_loss: 0.2224 - val_accuracy: 0.8550\n",
            "Epoch 76/100\n",
            "25/25 [==============================] - 0s 2ms/step - loss: 0.2054 - accuracy: 0.8825 - val_loss: 0.2164 - val_accuracy: 0.8700\n",
            "Epoch 77/100\n",
            "25/25 [==============================] - 0s 2ms/step - loss: 0.2031 - accuracy: 0.8838 - val_loss: 0.2179 - val_accuracy: 0.8650\n",
            "Epoch 78/100\n",
            "25/25 [==============================] - 0s 2ms/step - loss: 0.1987 - accuracy: 0.8863 - val_loss: 0.2104 - val_accuracy: 0.8750\n",
            "Epoch 79/100\n",
            "25/25 [==============================] - 0s 2ms/step - loss: 0.1960 - accuracy: 0.8900 - val_loss: 0.2073 - val_accuracy: 0.8750\n",
            "Epoch 80/100\n",
            "25/25 [==============================] - 0s 3ms/step - loss: 0.1915 - accuracy: 0.8900 - val_loss: 0.2081 - val_accuracy: 0.8700\n",
            "Epoch 81/100\n",
            "25/25 [==============================] - 0s 2ms/step - loss: 0.1874 - accuracy: 0.8950 - val_loss: 0.2010 - val_accuracy: 0.8700\n",
            "Epoch 82/100\n",
            "25/25 [==============================] - 0s 2ms/step - loss: 0.1833 - accuracy: 0.8975 - val_loss: 0.1942 - val_accuracy: 0.8750\n",
            "Epoch 83/100\n",
            "25/25 [==============================] - 0s 2ms/step - loss: 0.1887 - accuracy: 0.8888 - val_loss: 0.2078 - val_accuracy: 0.8700\n",
            "Epoch 84/100\n",
            "25/25 [==============================] - 0s 2ms/step - loss: 0.1796 - accuracy: 0.8975 - val_loss: 0.1889 - val_accuracy: 0.8750\n",
            "Epoch 85/100\n",
            "25/25 [==============================] - 0s 3ms/step - loss: 0.1757 - accuracy: 0.8988 - val_loss: 0.1786 - val_accuracy: 0.8900\n",
            "Epoch 86/100\n",
            "25/25 [==============================] - 0s 2ms/step - loss: 0.1716 - accuracy: 0.9025 - val_loss: 0.1787 - val_accuracy: 0.8850\n",
            "Epoch 87/100\n",
            "25/25 [==============================] - 0s 3ms/step - loss: 0.1698 - accuracy: 0.9038 - val_loss: 0.1787 - val_accuracy: 0.8900\n",
            "Epoch 88/100\n",
            "25/25 [==============================] - 0s 2ms/step - loss: 0.1660 - accuracy: 0.9050 - val_loss: 0.1726 - val_accuracy: 0.8900\n",
            "Epoch 89/100\n",
            "25/25 [==============================] - 0s 2ms/step - loss: 0.1637 - accuracy: 0.9050 - val_loss: 0.1693 - val_accuracy: 0.8900\n",
            "Epoch 90/100\n",
            "25/25 [==============================] - 0s 2ms/step - loss: 0.1604 - accuracy: 0.9087 - val_loss: 0.1706 - val_accuracy: 0.8900\n",
            "Epoch 91/100\n",
            "25/25 [==============================] - 0s 2ms/step - loss: 0.1581 - accuracy: 0.9087 - val_loss: 0.1650 - val_accuracy: 0.8900\n",
            "Epoch 92/100\n",
            "25/25 [==============================] - 0s 3ms/step - loss: 0.1532 - accuracy: 0.9112 - val_loss: 0.1647 - val_accuracy: 0.8850\n",
            "Epoch 93/100\n",
            "25/25 [==============================] - 0s 2ms/step - loss: 0.1508 - accuracy: 0.9112 - val_loss: 0.1699 - val_accuracy: 0.8900\n",
            "Epoch 94/100\n",
            "25/25 [==============================] - 0s 2ms/step - loss: 0.1487 - accuracy: 0.9125 - val_loss: 0.1565 - val_accuracy: 0.8900\n",
            "Epoch 95/100\n",
            "25/25 [==============================] - 0s 2ms/step - loss: 0.1456 - accuracy: 0.9137 - val_loss: 0.1542 - val_accuracy: 0.8950\n",
            "Epoch 96/100\n",
            "25/25 [==============================] - 0s 2ms/step - loss: 0.1425 - accuracy: 0.9175 - val_loss: 0.1465 - val_accuracy: 0.8950\n",
            "Epoch 97/100\n",
            "25/25 [==============================] - 0s 2ms/step - loss: 0.1349 - accuracy: 0.9463 - val_loss: 0.1364 - val_accuracy: 0.9600\n",
            "Epoch 98/100\n",
            "25/25 [==============================] - 0s 2ms/step - loss: 0.1260 - accuracy: 0.9375 - val_loss: 0.1287 - val_accuracy: 0.9650\n",
            "Epoch 99/100\n",
            "25/25 [==============================] - 0s 3ms/step - loss: 0.0881 - accuracy: 0.9725 - val_loss: 0.0622 - val_accuracy: 0.9750\n",
            "Epoch 100/100\n",
            "25/25 [==============================] - 0s 2ms/step - loss: 0.0616 - accuracy: 0.9775 - val_loss: 0.0770 - val_accuracy: 0.9700\n"
          ]
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "Our model performed pretty well with `Accuracy` **99%**.\n",
        "\n",
        "Now to replicate the figure we need to plot the decision boundary. Let's do then,"
      ],
      "metadata": {
        "id": "Wxe2FWcnEYmR"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "def plot_decision_boundary(model, X, y):\n",
        "  # Define the axis boundaries of the plot and create a meshgrid\n",
        "  x_min, x_max = X[:, 0].min() - 0.1, X[:, 0].max() + 0.1\n",
        "  y_min, y_max = X[:, 1].min() - 0.1, X[:, 1].max() + 0.1\n",
        "  xx, yy = np.meshgrid(np.linspace(x_min, x_max, 100),\n",
        "                       np.linspace(y_min, y_max, 100))\n",
        "  \n",
        "  # Create X values (we're going to predict on all of these)\n",
        "  x_in = np.c_[xx.ravel(), yy.ravel()] # stack 2D arrays together\n",
        "\n",
        "  # Make predictions using trained model\n",
        "  y_pred = model.predict(x_in)\n",
        "\n",
        "  # Check for multi-class\n",
        "  if model.output_shape[-1] > 1: \n",
        "    print(\"Doing multi-class classification...\")\n",
        "    # We have to reshape our predictions to get them ready for plotting\n",
        "    y_pred = np.argmax(y_pred, axis=1).reshape(xx.shape)\n",
        "  else:\n",
        "    print(\"Doing binary-class classification...\")\n",
        "    y_pred = np.round(np.max(y_pred, axis=1)).reshape(xx.shape)\n",
        "  \n",
        "  # plot decision boundary\n",
        "  plt.contourf(xx, yy, y_pred, cmap=plt.cm.RdYlBu, alpha = 0.7)\n",
        "  plt.scatter(X[:,0], X[:,1], c = y, s=40, cmap=plt.cm.RdYlBu)\n",
        "  plt.xlim(xx.min(), xx.max())\n",
        "  plt.ylim(yy.min(), yy.max())"
      ],
      "metadata": {
        "id": "yIoUkN1rEwC0"
      },
      "execution_count": 11,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "plot_decision_boundary(model_1, X, y)"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 300
        },
        "id": "RH5BO1rrF9lJ",
        "outputId": "0fca835a-6a42-46de-e04b-ccc45e2e7651"
      },
      "execution_count": 12,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "313/313 [==============================] - 0s 907us/step\n",
            "Doing binary-class classification...\n"
          ]
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 432x288 with 1 Axes>"
            ],
            "image/png": "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\n"
          },
          "metadata": {
            "needs_background": "light"
          }
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "There is another way of plotting decision boundaries using `mlxtend.plotting`.\n",
        "\n",
        "Lets see..."
      ],
      "metadata": {
        "id": "6dWBzHXnI-cX"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "from mlxtend.plotting import plot_decision_regions\n"
      ],
      "metadata": {
        "id": "AsjPoTRR17oo"
      },
      "execution_count": 24,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "plot_decision_regions(X_test, y_test, clf=model_1,legend=2)\n",
        "plt.show();"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 339
        },
        "id": "Rr4hAMHi6xRd",
        "outputId": "a2c3246b-991d-45f3-81a9-1d91f1a3a6a5"
      },
      "execution_count": 29,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "3888/3888 [==============================] - 4s 973us/step\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "/usr/local/lib/python3.7/dist-packages/mlxtend/plotting/decision_regions.py:244: MatplotlibDeprecationWarning: Passing unsupported keyword arguments to axis() will raise a TypeError in 3.3.\n",
            "  ax.axis(xmin=xx.min(), xmax=xx.max(), y_min=yy.min(), y_max=yy.max())\n"
          ]
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 432x288 with 1 Axes>"
            ],
            "image/png": "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\n"
          },
          "metadata": {
            "needs_background": "light"
          }
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "## Create a classification dataset using Scikit-Learns `make_moons()`"
      ],
      "metadata": {
        "id": "6b12_bDL7gns"
      }
    },
    {
      "cell_type": "markdown",
      "source": [
        "Create a classification dataset using Scikit-learns [make_moons()](https://scikit-learn.org/stable/modules/generated/sklearn.datasets.make_moons.html) function, visualize it and then build a model to fit it at over 85% accuracy."
      ],
      "metadata": {
        "id": "Vkdw6mM98Gur"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "# import the make_moons() dataset\n",
        "from sklearn.datasets import make_moons\n",
        "n_samples = 1000\n",
        "\n",
        "X, y = make_moons(n_samples)"
      ],
      "metadata": {
        "id": "RfrkM4N18Zva"
      },
      "execution_count": 30,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "# view the data\n",
        "X[:10], y[:10]"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "vn7KhN2w8ynf",
        "outputId": "e2a34c3e-c011-4abf-adc3-ce5f83422aa0"
      },
      "execution_count": 31,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "(array([[ 0.32125705, -0.23437593],\n",
              "        [ 0.75997907, -0.47076771],\n",
              "        [-0.98663265,  0.16296018],\n",
              "        [ 0.25503244, -0.16710069],\n",
              "        [ 1.99611811,  0.4119732 ],\n",
              "        [ 0.02832721,  0.9995987 ],\n",
              "        [ 0.40358981, -0.30267982],\n",
              "        [-0.95468484,  0.29761864],\n",
              "        [ 0.73223568,  0.68105132],\n",
              "        [-0.44640126,  0.8948329 ]]), array([1, 1, 0, 1, 1, 0, 1, 0, 0, 0]))"
            ]
          },
          "metadata": {},
          "execution_count": 31
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "X.shape, y.shape"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "O6trNGcb83Jn",
        "outputId": "9fc4d21a-0c59-47fa-dbb1-e6e7273e618d"
      },
      "execution_count": 32,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "((1000, 2), (1000,))"
            ]
          },
          "metadata": {},
          "execution_count": 32
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "# Split the data first into training and test samples\n",
        "X_train, X_test = X[:750], X[750:]\n",
        "y_train, y_test = y[:750], y[750:]"
      ],
      "metadata": {
        "id": "sH7QzPg9858w"
      },
      "execution_count": 33,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "# plot the train data into scatter plot\n",
        "plt.scatter(X_train[:, 0], X_train[:, 1], c=y_train);"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 265
        },
        "id": "DJ3XmjIe9Sum",
        "outputId": "e8cb18aa-745e-4445-d941-8c4cedd53c18"
      },
      "execution_count": 35,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 432x288 with 1 Axes>"
            ],
            "image/png": "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\n"
          },
          "metadata": {
            "needs_background": "light"
          }
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "Data looks non-linear, so we will need activation functions which supports non-linearity.\n",
        "\n",
        "Let's do the modelling now,"
      ],
      "metadata": {
        "id": "trk6cn999h5t"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "# set the random seed\n",
        "tf.random.set_seed(17)\n",
        "\n",
        "# create a model\n",
        "model_2 = tf.keras.Sequential([\n",
        "    tf.keras.layers.Dense(4, activation='relu'),\n",
        "    tf.keras.layers.Dense(4, activation='relu'),\n",
        "    tf.keras.layers.Dense(1, activation='sigmoid') # output layer, binary classification so sigmoid activation\n",
        "])\n",
        "\n",
        "# compile a model\n",
        "model_2.compile(loss='binary_crossentropy',\n",
        "                optimizer='Adam',\n",
        "                metrics=['accuracy'])\n",
        "\n",
        "# fit the model\n",
        "history_2 = model_2.fit(X_train,\n",
        "                        y_train,\n",
        "                        epochs=50)"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "vO28cglw92r9",
        "outputId": "87fcd0c9-5641-44f8-d10d-655e0ccbbf81"
      },
      "execution_count": 36,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Epoch 1/50\n",
            "24/24 [==============================] - 1s 2ms/step - loss: 0.6382 - accuracy: 0.7400\n",
            "Epoch 2/50\n",
            "24/24 [==============================] - 0s 3ms/step - loss: 0.6233 - accuracy: 0.8093\n",
            "Epoch 3/50\n",
            "24/24 [==============================] - 0s 2ms/step - loss: 0.6003 - accuracy: 0.8307\n",
            "Epoch 4/50\n",
            "24/24 [==============================] - 0s 2ms/step - loss: 0.5745 - accuracy: 0.8320\n",
            "Epoch 5/50\n",
            "24/24 [==============================] - 0s 1ms/step - loss: 0.5518 - accuracy: 0.8307\n",
            "Epoch 6/50\n",
            "24/24 [==============================] - 0s 1ms/step - loss: 0.5307 - accuracy: 0.8360\n",
            "Epoch 7/50\n",
            "24/24 [==============================] - 0s 1ms/step - loss: 0.5126 - accuracy: 0.8373\n",
            "Epoch 8/50\n",
            "24/24 [==============================] - 0s 1ms/step - loss: 0.4965 - accuracy: 0.8413\n",
            "Epoch 9/50\n",
            "24/24 [==============================] - 0s 1ms/step - loss: 0.4830 - accuracy: 0.8427\n",
            "Epoch 10/50\n",
            "24/24 [==============================] - 0s 1ms/step - loss: 0.4702 - accuracy: 0.8533\n",
            "Epoch 11/50\n",
            "24/24 [==============================] - 0s 1ms/step - loss: 0.4584 - accuracy: 0.8600\n",
            "Epoch 12/50\n",
            "24/24 [==============================] - 0s 1ms/step - loss: 0.4474 - accuracy: 0.8627\n",
            "Epoch 13/50\n",
            "24/24 [==============================] - 0s 1ms/step - loss: 0.4373 - accuracy: 0.8613\n",
            "Epoch 14/50\n",
            "24/24 [==============================] - 0s 1ms/step - loss: 0.4275 - accuracy: 0.8587\n",
            "Epoch 15/50\n",
            "24/24 [==============================] - 0s 2ms/step - loss: 0.4178 - accuracy: 0.8573\n",
            "Epoch 16/50\n",
            "24/24 [==============================] - 0s 1ms/step - loss: 0.4087 - accuracy: 0.8573\n",
            "Epoch 17/50\n",
            "24/24 [==============================] - 0s 1ms/step - loss: 0.3997 - accuracy: 0.8573\n",
            "Epoch 18/50\n",
            "24/24 [==============================] - 0s 1ms/step - loss: 0.3906 - accuracy: 0.8587\n",
            "Epoch 19/50\n",
            "24/24 [==============================] - 0s 1ms/step - loss: 0.3820 - accuracy: 0.8573\n",
            "Epoch 20/50\n",
            "24/24 [==============================] - 0s 1ms/step - loss: 0.3735 - accuracy: 0.8573\n",
            "Epoch 21/50\n",
            "24/24 [==============================] - 0s 1ms/step - loss: 0.3652 - accuracy: 0.8560\n",
            "Epoch 22/50\n",
            "24/24 [==============================] - 0s 1ms/step - loss: 0.3574 - accuracy: 0.8573\n",
            "Epoch 23/50\n",
            "24/24 [==============================] - 0s 1ms/step - loss: 0.3497 - accuracy: 0.8600\n",
            "Epoch 24/50\n",
            "24/24 [==============================] - 0s 1ms/step - loss: 0.3423 - accuracy: 0.8613\n",
            "Epoch 25/50\n",
            "24/24 [==============================] - 0s 1ms/step - loss: 0.3355 - accuracy: 0.8613\n",
            "Epoch 26/50\n",
            "24/24 [==============================] - 0s 1ms/step - loss: 0.3287 - accuracy: 0.8653\n",
            "Epoch 27/50\n",
            "24/24 [==============================] - 0s 1ms/step - loss: 0.3218 - accuracy: 0.8667\n",
            "Epoch 28/50\n",
            "24/24 [==============================] - 0s 1ms/step - loss: 0.3156 - accuracy: 0.8693\n",
            "Epoch 29/50\n",
            "24/24 [==============================] - 0s 1ms/step - loss: 0.3095 - accuracy: 0.8720\n",
            "Epoch 30/50\n",
            "24/24 [==============================] - 0s 1ms/step - loss: 0.3036 - accuracy: 0.8747\n",
            "Epoch 31/50\n",
            "24/24 [==============================] - 0s 2ms/step - loss: 0.2980 - accuracy: 0.8773\n",
            "Epoch 32/50\n",
            "24/24 [==============================] - 0s 2ms/step - loss: 0.2925 - accuracy: 0.8813\n",
            "Epoch 33/50\n",
            "24/24 [==============================] - 0s 1ms/step - loss: 0.2873 - accuracy: 0.8840\n",
            "Epoch 34/50\n",
            "24/24 [==============================] - 0s 1ms/step - loss: 0.2820 - accuracy: 0.8867\n",
            "Epoch 35/50\n",
            "24/24 [==============================] - 0s 1ms/step - loss: 0.2771 - accuracy: 0.8880\n",
            "Epoch 36/50\n",
            "24/24 [==============================] - 0s 1ms/step - loss: 0.2722 - accuracy: 0.8920\n",
            "Epoch 37/50\n",
            "24/24 [==============================] - 0s 1ms/step - loss: 0.2677 - accuracy: 0.8920\n",
            "Epoch 38/50\n",
            "24/24 [==============================] - 0s 1ms/step - loss: 0.2630 - accuracy: 0.8960\n",
            "Epoch 39/50\n",
            "24/24 [==============================] - 0s 1ms/step - loss: 0.2583 - accuracy: 0.8973\n",
            "Epoch 40/50\n",
            "24/24 [==============================] - 0s 1ms/step - loss: 0.2538 - accuracy: 0.8987\n",
            "Epoch 41/50\n",
            "24/24 [==============================] - 0s 1ms/step - loss: 0.2495 - accuracy: 0.9000\n",
            "Epoch 42/50\n",
            "24/24 [==============================] - 0s 1ms/step - loss: 0.2459 - accuracy: 0.9027\n",
            "Epoch 43/50\n",
            "24/24 [==============================] - 0s 1ms/step - loss: 0.2430 - accuracy: 0.9040\n",
            "Epoch 44/50\n",
            "24/24 [==============================] - 0s 1ms/step - loss: 0.2407 - accuracy: 0.9027\n",
            "Epoch 45/50\n",
            "24/24 [==============================] - 0s 1ms/step - loss: 0.2383 - accuracy: 0.9027\n",
            "Epoch 46/50\n",
            "24/24 [==============================] - 0s 1ms/step - loss: 0.2365 - accuracy: 0.9027\n",
            "Epoch 47/50\n",
            "24/24 [==============================] - 0s 1ms/step - loss: 0.2346 - accuracy: 0.9040\n",
            "Epoch 48/50\n",
            "24/24 [==============================] - 0s 1ms/step - loss: 0.2328 - accuracy: 0.9013\n",
            "Epoch 49/50\n",
            "24/24 [==============================] - 0s 1ms/step - loss: 0.2312 - accuracy: 0.9013\n",
            "Epoch 50/50\n",
            "24/24 [==============================] - 0s 1ms/step - loss: 0.2299 - accuracy: 0.9013\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "# let's evaluate the model\n",
        "model_2.evaluate(X_test, y_test)"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "L0vxNtiE-oNB",
        "outputId": "3a7e2f03-2d34-4d41-f012-e5c23e4afaac"
      },
      "execution_count": 37,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "8/8 [==============================] - 0s 1ms/step - loss: 0.2316 - accuracy: 0.8960\n"
          ]
        },
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "[0.2315957397222519, 0.8960000276565552]"
            ]
          },
          "metadata": {},
          "execution_count": 37
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "So on testing samples, model's accuracy was **89%** which was our goal."
      ],
      "metadata": {
        "id": "HSPSBmzq-vjH"
      }
    },
    {
      "cell_type": "markdown",
      "source": [
        "## Recreate TensorFlow's softmax activation function in your code.\n",
        "\n",
        "Recreate [TensorFlow's Softmax Activation](https://en.wikipedia.org/wiki/Softmax_function) function in your own code. Make sure it can accept a tensor and return that tensor after having softmax function applied to it."
      ],
      "metadata": {
        "id": "eFv2swYa-64j"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "# Let's create a tensor with random numbers ranging from -1 to 1\n",
        "random = tf.constant(tf.range(-1, 1, 0.1))"
      ],
      "metadata": {
        "id": "lCECAbOBBD3k"
      },
      "execution_count": 38,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "source": [
        "So what is softmax function ?\n",
        "\n",
        "It basically return the max between 0 and the passed value, if some passed value is less than 0 then it treat as 0 and if some value is greater than 0 then it treats with original value.\n",
        "In short max(0, n)"
      ],
      "metadata": {
        "id": "g6LDHhG-BtZZ"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "random"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "rvuwRVDHBuOg",
        "outputId": "1f42fb4d-062b-49e2-dfe4-6e66a56b9620"
      },
      "execution_count": 40,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "<tf.Tensor: shape=(20,), dtype=float32, numpy=\n",
              "array([-1.00000000e+00, -8.99999976e-01, -7.99999952e-01, -6.99999928e-01,\n",
              "       -5.99999905e-01, -4.99999911e-01, -3.99999917e-01, -2.99999923e-01,\n",
              "       -1.99999928e-01, -9.99999270e-02,  7.45058060e-08,  1.00000076e-01,\n",
              "        2.00000077e-01,  3.00000072e-01,  4.00000066e-01,  5.00000060e-01,\n",
              "        6.00000083e-01,  7.00000107e-01,  8.00000131e-01,  9.00000155e-01],\n",
              "      dtype=float32)>"
            ]
          },
          "metadata": {},
          "execution_count": 40
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "so after max()"
      ],
      "metadata": {
        "id": "-16VE9j5CPN4"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "tf.maximum(0, random)"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "Kc3kn9PVCTUu",
        "outputId": "4227ad86-a719-40d7-99bb-e1854c006a51"
      },
      "execution_count": 44,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "<tf.Tensor: shape=(20,), dtype=float32, numpy=\n",
              "array([0.00000000e+00, 0.00000000e+00, 0.00000000e+00, 0.00000000e+00,\n",
              "       0.00000000e+00, 0.00000000e+00, 0.00000000e+00, 0.00000000e+00,\n",
              "       0.00000000e+00, 0.00000000e+00, 7.45058060e-08, 1.00000076e-01,\n",
              "       2.00000077e-01, 3.00000072e-01, 4.00000066e-01, 5.00000060e-01,\n",
              "       6.00000083e-01, 7.00000107e-01, 8.00000131e-01, 9.00000155e-01],\n",
              "      dtype=float32)>"
            ]
          },
          "metadata": {},
          "execution_count": 44
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "So if we plot it using `matplotlib`"
      ],
      "metadata": {
        "id": "JU7Y0riFCYB9"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "plt.plot(tf.maximum(0, random));"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 265
        },
        "id": "kfsBosWoCrdM",
        "outputId": "9388f975-2ee4-4ad6-ca75-fa10c7be7799"
      },
      "execution_count": 46,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 432x288 with 1 Axes>"
            ],
            "image/png": "iVBORw0KGgoAAAANSUhEUgAAAXQAAAD4CAYAAAD8Zh1EAAAABHNCSVQICAgIfAhkiAAAAAlwSFlzAAALEgAACxIB0t1+/AAAADh0RVh0U29mdHdhcmUAbWF0cGxvdGxpYiB2ZXJzaW9uMy4yLjIsIGh0dHA6Ly9tYXRwbG90bGliLm9yZy+WH4yJAAAdaUlEQVR4nO3deXxU1f3/8deHkBDZQQKyL7LJIgQjglp3EawVl2JDN617+wXB3VaLVlt/Vau1ULtotYtfTdhcsGJxo9YNNRISCGvYwxr2NZDlfP/I0F+MCRnIzJxZ3s/HI4/cuffM3Pfj5s5nTs7MmWvOOUREJPY18B1ARERCQwVdRCROqKCLiMQJFXQRkTihgi4iEica+tpxmzZtXLdu3XztXkQkJn355ZfbnHNpNW3zVtC7detGTk6Or92LiMQkM1tb2zYNuYiIxAkVdBGROKGCLiISJ1TQRUTihAq6iEicUEEXEYkTKugiInFCBV1EJEIqKhy//Odi1m0/EJbHV0EXEYmQZz9cxV8+Ws0nK7eF5fFV0EVEIiB33U5+M2cZ3xzYnu+c3jks+1BBFxEJsz0lpdyWnUu75qk8etVAzCws+/H2XS4iIonAOcf9ry5i464Spt0ynBYnJIdtX+qhi4iE0fQvi3gjbyN3XNyb07q2Cuu+VNBFRMKkcOs+Hny9gOE9TuTWc08O+/5U0EVEwqCktJzxWbmckJLE05mDSWoQnnHzqjSGLiISBr9+aylLNu3hhesyaNc8NSL7VA9dRCTE3l28hb99sobrz+rOBX3bRWy/KugiIiG0eXcJd8/Io3+H5tw7qk9E962CLiISIuUVjolTczlUVsGUsek0apgU0f1rDF1EJET+MLeQeat28Jsxg+iR1jTi+1cPXUQkBHLW7ODp91YwenAHrh7S0UsGFXQRkXrafaCUCdkL6NTqBH55xYCwTe2vi4ZcRETqwTnHfa/ks2VPCTN/fCbNUsM3tb8u6qGLiNTDy5+v461Fm7n7kj4M6tzSaxYVdBGR47Rs814efmMx5/RO46Zv9PAdRwVdROR4VE7tn0+z1GSeHDOIBhGY2l8XjaGLiByHR/65mOVb9vGP64eS1qyR7zhAkD10MxtpZsvMrNDM7qthexczm2tmuWaWb2aXhj6qiEh0eGvhJl76bB23nNODc3qn+Y7zX3UWdDNLAp4BRgH9gLFm1q9asweAac65dCAT+EOog4qIRIOinQe4d2Y+gzq14M4RkZ3aX5dgeuhDgULn3Crn3GEgGxhdrY0DmgeWWwAbQxdRRCQ6lJVXMDF7ARUOJo9NJ6VhdL0NGUyajsD6KreLAuuqegj4vpkVAbOB8TU9kJndbGY5ZpZTXFx8HHFFRPyZ/N4Kctbu5FdXDqDriU18x/maUL28jAX+5pzrBFwKvGhmX3ts59yzzrkM51xGWlr0jDuJiNTl05XbmTK3kG+f1onRg/1M7a9LMAV9A9C5yu1OgXVV3QBMA3DOfQqkAm1CEVBExLcd+w8zcWou3ds04ReX9/cdp1bBFPQvgF5m1t3MUqh803NWtTbrgAsBzOwUKgu6xlREJOY557hnRh4795cyOTOdJo2i99PedRZ051wZMA6YAyyh8tMsBWb2sJldHmh2J3CTmeUBWcB1zjkXrtAiIpHy90/W8O6Srdw3qi8DOrbwHeeognqpcc7NpvLNzqrrJlVZXgycFdpoIiJ+FWzczaOzl3Jh37b86KxuvuPUKbo+cyMiEiUOHC5jfFYuLRsn88SYQd6+EvdYRO9gkIiIRw/NKmD1tv28dOMZtG6S4jtOUNRDFxGpZlbeRqblFPE/5/XkzJNj5wN7KugiIlWs236A+19ZyGldWzHxol6+4xwTFXQRkYDS8gpuy84Fg99lDqZhUmyVSI2hi4gEPPn2chas38UfvjeETq0a+45zzGLr5UdEJEw+XFHMnz5YydihXbh0YHvfcY6LCrqIJLxt+w5xx7Q8erVtyqTLqn87eOzQkIuIJLSKCsed0/LYc7CUF28YygkpSb4jHTf10EUkoT3/0Wo+WF7MA5f1o+9Jzeu+QxRTQReRhJVftIvH5yzlkv7t+P4ZXXzHqTcVdBFJSPsOlXFbVi5pTRvx2NWnxsTU/rpoDF1EEtLPX1vEuh0HyL55OC0bx8bU/rqohy4iCeeV+UW8mruBCRf2Zmj31r7jhIwKuogklNXb9vPAa4sY2r014y7o6TtOSKmgi0jCOFRWzvis+aQ0bMDvMgeT1CD2x82r0hi6iCSMx/+1jEUb9vDsD06jfYsTfMcJOfXQRSQhzF26lec/Ws21w7syov9JvuOEhQq6iMS9rXtKuGt6Hn1PasZPLz3Fd5ywUUEXkbhWUeG4fdoCDhwu5/ffTSc1OXan9tdFY+giEtf++MFKPi7czmNXD6Rn22a+44SVeugiErfmr9vJU+8s55untueajM6+44SdCrqIxKXdB0u5LSuX9i1S+X9XDYyLqf110ZCLiMQd5xw/e3Uhm3aXMP3W4TRPTfYdKSLUQxeRuDMtZz1v5m/ijot7M6RLK99xIkYFXUTiSuHWvTw4q4Czep7Ij8892XeciFJBF5G4UVJazriXc2mc0pDfXjOYBnE2tb8uGkMXkbjx6OwlLN28l79edzptm6f6jhNx6qGLSFx4u2Az//h0LTec3Z3z+7b1HccLFXQRiXmbdh/knpn5DOjYnHtG9vEdxxsVdBGJaeUVjgnZCzhcVsHkzHQaNYzfqf110Ri6iMS0Ke+v4PPVO3hyzCB6pDX1Hccr9dBFJGZ9vnoHk99bwZXpHbn6tE6+43ingi4iMWnXgcNMzM6lc+vGPHLFAN9xooKGXEQk5jjnuGdGPsX7DjHzx2fStJFKGQTZQzezkWa2zMwKzey+WtpcY2aLzazAzF4ObUwRkf/vf+et5e3FW7jnkr6c2qml7zhRo86XNTNLAp4BLgaKgC/MbJZzbnGVNr2AnwJnOed2mllifghURMJu6eY9PPLmEs7pncYNZ3f3HSeqBNNDHwoUOudWOecOA9nA6GptbgKecc7tBHDObQ1tTBEROHi4nPEv59I8NZknxwxKuKn9dQmmoHcE1le5XRRYV1VvoLeZfWxm88xsZE0PZGY3m1mOmeUUFxcfX2IRSVgP/7OAFVv38dQ1g0hr1sh3nKgTqk+5NAR6AecBY4HnzOxrA1vOuWedcxnOuYy0tLQQ7VpEEsGb+ZvI+nw9t5zbg3N6q37UJJiCvgGoeu2mToF1VRUBs5xzpc651cByKgu8iEi9rd9xgPteyWdQ55bcNSJxp/bXJZiC/gXQy8y6m1kKkAnMqtbmNSp755hZGyqHYFaFMKeIJKjS8gomZOeCgymZ6SQnafpMbeo8Ms65MmAcMAdYAkxzzhWY2cNmdnmg2Rxgu5ktBuYCdzvntocrtIgkjqffXc78dbv45ZUD6HJiY99xolpQn8Z3zs0GZldbN6nKsgPuCPyIiITEJ4Xb+MO/VzLmtE6MHlz9sxhSnf53EZGotH3fISZOXUD3Nk34xej+vuPEBBV0EYk6zjnunpHPrgOlTBmbTuMUTe0Phgq6iESdv368hveXbuWnl/alf4cWvuPEDBV0EYkqizbs5tdvLeWiU9py3ZndfMeJKSroIhI19h8qY3xWLq2aJPP4twdhpqn9x0IDUyISNSa9XsCa7ft5+cZhtG6S4jtOzFEPXUSiwmu5G5g5v4hx5/dk+Mkn+o4Tk1TQRcS7tdv388Bri8jo2ooJF+pbQ46XCrqIeHW4rILxWbk0MHg6czANNbX/uGkMXUS8evLtZeQX7eaP3xtCp1aa2l8feikUEW8+WF7Mn/+ziu+e0YVRA9v7jhPzVNBFxIute0u4c9oCerdryqTL+vmOExc05CIiEVdR4bhzWh57S8p4+aZhpCYn+Y4UF9RDF5GIe+7DVXy4YhuTvtWP3u2a+Y4TN1TQRSSiFqzfxRNzljFqwEl8d2gX33Hiigq6iETM3pJSbsvKpV3zVH591ama2h9iGkMXkYhwznH/q4vYsOsgU28eRovGyb4jxR310EUkImZ8WcSsvI1MvLAXGd1a+44Tl1TQRSTsVhbvY9LrBQzr0ZqfnN/Td5y4pYIuImF1qKyc8S/nkprcgKe/k05SA42bh4vG0EUkrH791lIWb9rDX36YwUktUn3HiWvqoYtI2Ly3ZAt//XgN153ZjYv6tfMdJ+6poItIWGzZU8Jd0/M4pX1z7hvV13echKCCLiIhV17hmJi9gJLSCn7/3XRN7Y8QjaGLSMj98d+FfLpqO49/+1ROTmvqO07CUA9dRELqy7U7+O27K/jWoA6MOa2T7zgJRQVdREJm98FSbstaQIeWqfzqygGa2h9hGnIRkZBwzvHTV/LZsqeE6bcOp3mqpvZHmnroIhIS2V+sZ/bCzdw5og/pXVr5jpOQVNBFpN5WbNnLL94o4Bu92nDLOT18x0lYKugiUi8lpeWMezmXJikNefKaQTTQ1H5vNIYuIvXyqzeXsGzLXv72o9Np20xT+31SD11Ejtu/Fm3mxXlruekb3TmvT1vfcRKeCrqIHJcNuw5y78x8Tu3Ugrsv0dT+aKCCLiLHrKy8gtuzF1BWXsHkzHRSGqqURIOg/gpmNtLMlplZoZndd5R2V5uZM7OM0EUUkWgz5f1CPl+zg19eOYBubZr4jiMBdRZ0M0sCngFGAf2AsWbWr4Z2zYAJwGehDiki0WPequ1MeX8FVw3pyJXpmtofTYLpoQ8FCp1zq5xzh4FsYHQN7R4BHgNKQphPRKLIrgOHuX3qArqe2ISHRw/wHUeqCaagdwTWV7ldFFj3X2Y2BOjsnHvzaA9kZjebWY6Z5RQXFx9zWBHxxznH3TPy2bbvEJMz02naSJ96jjb1fifDzBoATwF31tXWOfescy7DOZeRlpZW312LSAS9OG8t7yzewr0j+zKwUwvfcaQGwRT0DUDnKrc7BdYd0QwYAPzbzNYAw4BZemNUJH4s2bSHX765hPP7pHHD2d19x5FaBFPQvwB6mVl3M0sBMoFZRzY653Y759o457o557oB84DLnXM5YUksIhF14HAZ47NyaXFCMr8ZM0hfiRvF6izozrkyYBwwB1gCTHPOFZjZw2Z2ebgDiohfD7+xmJXF+3j6O4M5sWkj33HkKIJ6V8M5NxuYXW3dpFranlf/WCISDd7I20j2F+v5yXknc1bPNr7jSB00vUtEarR+xwF+9spC0ru05PaLe/uOI0FQQReRryktr+C27FwAJmemk5ykUhEL9EFSEfma376znNx1u/j9d9Pp3Lqx7zgSJL3sishXfFy4jT9+sJLM0ztz2akdfMeRY6CCLiL/tX3fISZOXUCPNk2Y9K2vfWWTRDkNuYgIABUVjrum57H7YCn/uH4ojVNUHmKNeugiAsALH69m7rJiHvjmKZzSvrnvOHIcVNBFhIVFu3nsX0u5uF87fjCsq+84cpxU0EUS3L5DZYzPmk+bpo14/OpTNbU/hmmQTCTBTXp9Eet2HCDrpmG0apLiO47Ug3roIgns1dwiXpm/gfEX9OKMHif6jiP1pIIukqBWb9vPA68uYmi31oy/oKfvOBICKugiCehwWQW3ZeXSMKkBT2cOpqGm9scFjaGLJKAn5ixl4Ybd/PkHp9Gh5Qm+40iI6GVZJMHMXbaV5z5czQ+GdeWS/if5jiMhpIIukkC27i3hrml59D2pGfd/8xTfcSTENOQikiAqKhx3TM1j/+EysscOIzU5yXckCTH10EUSxJ//s4qPCrfx4Lf606tdM99xJAxU0EUSQO66nTz59jK+ObA9mad39h1HwkQFXSTO7Skp5bbsXNo1T+XRqwZqan8c0xi6SBxzznH/q4vYuKuEabcMp8UJyb4jSRiphy4Sx6bnFPFG3kbuuLg3p3Vt5TuOhJkKukicKty6jwdnFTC8x4nceu7JvuNIBKigi8ShktJyxmflckJKEk9nDiapgcbNE4HG0EXi0K/fWsqSTXt44boM2jVP9R1HIkQ9dJE4887iLfztkzVcf1Z3LujbzncciSAVdJE4snl3CXfPyKN/h+bcO6qP7zgSYSroInGivMIxITuXw2UVTBmbTqOGmtqfaDSGLhInnplbyGerd/CbMYPokdbUdxzxQD10kTiQs2YHT7+7nCsGd+DqIR19xxFPVNBFYtzuA6VMyF5A59aNeeSKAZran8A05CISw5xz3Dszny17Spj54zNplqqp/YlMPXSRGPbSZ+v4V8Fm7hnZh0GdW/qOI56poIvEqGWb9/LIPxdzTu80bjy7h+84EgVU0EVi0MHD5YzPmk+z1GSeHDOIBpraLwRZ0M1spJktM7NCM7uvhu13mNliM8s3s/fMrGvoo4rIEY+8uZjlW/bx1DWDSGvWyHcciRJ1FnQzSwKeAUYB/YCxZtavWrNcIMM5dyowA3g81EFFpNJbCzfx8mfruOXcHpzTO813HIkiwfTQhwKFzrlVzrnDQDYwumoD59xc59yBwM15QKfQxhQRgA27DnLvzHwGdW7JXSM0tV++KpiC3hFYX+V2UWBdbW4A3qppg5ndbGY5ZpZTXFwcfEoRoay8gglZuVQ4mJw5mOQkvQUmXxXSM8LMvg9kAE/UtN0596xzLsM5l5GWpn8VRY7F795bQc7anfzqygF0PbGJ7zgShYKZWLQBqHqZ8E6BdV9hZhcB9wPnOucOhSaeiAB8unI7v59byJjTOjF6sKb2S82C6aF/AfQys+5mlgJkArOqNjCzdODPwOXOua2hjymSuHbsP8zEqbl0b9OEhy7v7zuORLE6C7pzrgwYB8wBlgDTnHMFZvawmV0eaPYE0BSYbmYLzGxWLQ8nIsfAOcc9M/LYub+UKWPTadJI39YhtQvq7HDOzQZmV1s3qcryRSHOJSLA3z9Zw7tLtvLgt/rRv0ML33EkyultcpEoVbBxN4/OXsqFfdty3ZndfMeRGKCCLhKFDhwuY3xWLq2aJPPEmEH6SlwJigbkRKLQQ7MKWL1tPy/deAatm6T4jiMxQj10kSgzK28j03KKGHd+T848uY3vOBJDVNBFosi67Qf42SsLOa1rKyZc2Mt3HIkxKugiUaK0vILx2bk0MPhd5mAaamq/HCONoYtEiSffXk7e+l384XtD6NSqse84EoPUBRCJAh+uKOZPH6xk7NAuXDqwve84EqNU0EU8K957iNun5tG7XVMmXVb9UgMiwdOQi4hHFRWOu6bnsbeklJduPIMTUpJ8R5IYph66iEfPf7SaD5YX8/PL+tHnpGa+40iMU0EX8SS/aBePz1nKJf3b8b0zuviOI3FABV3Eg70lpYzPyiWtaSMeu/pUTe2XkNAYuogHk14vYP2OA0y9ZTgtG2tqv4SGeugiETbzyyJezd3AhAt7c3q31r7jSBxRQReJoNXb9vPz1xdxRvfWjLugp+84EmdU0EUi5FBZOeNenk9KwwY8nTmYpAYaN5fQ0hi6SIQ8/q9lFGzcw3M/zKB9ixN8x5E4pB66SATMXbqV5z9azbXDu3Jxv3a+40icUkEXCbMte0q4c3oep7Rvzk8vPcV3HIljKugiYVRe4bh96gIOHi5nyth0UpM1tV/CR2PoImH0pw9W8snK7Tx29UB6tm3qO47EOfXQRcLky7U7eeqd5Vx2anuuyejsO44kABV0kTDYfbCU27Jyad8ilUevGqip/RIRGnIRCTHnHD97ZSGb95Qw/dbhNE9N9h1JEoR66CIhNvWL9by5cBN3jujNkC6tfMeRBKKCLhJCK7bs5aE3Cji7ZxtuPedk33Ekwaigi4RISWk547NyaZLSkKeuGUQDTe2XCNMYukiIPDp7CUs37+WvPzqdts1TfceRBKQeukgIzCnYzD8+XcuNZ3fn/D5tfceRBKWCLlJPG3cd5J4Z+Qzs2IJ7Rvb1HUcSmAq6SD2UVzgmZi+grLyCyWPTSWmop5T4ozF0kXqY8v4KPl+zg6euGUT3Nk18x5EEp+6EyHH6bNV2Jr+3gqvSO3LVkE6+44iooIscj537DzNx6gK6tG7Mw1cM8B1HBAiyoJvZSDNbZmaFZnZfDdsbmdnUwPbPzKxbqIOKRAvnHPfOzGfbvkNMGTuEpo00cinRoc6CbmZJwDPAKKAfMNbM+lVrdgOw0znXE/gt8Fiog4pEi/+dt5a3F2/h3pF9Gdiphe84Iv8VTNdiKFDonFsFYGbZwGhgcZU2o4GHAsszgN+bmTnnXAizAjDti/U89+GqUD+sSNDWbN/PeX3SuP6s7r6jiHxFMAW9I7C+yu0i4Iza2jjnysxsN3AisK1qIzO7GbgZoEuXLscVuGXjZHq104UCxJ+Mbq24a0QfTe2XqBPRwT/n3LPAswAZGRnH1Xsf0f8kRvQ/KaS5RETiQTBvim4Aql5upVNgXY1tzKwh0ALYHoqAIiISnGAK+hdALzPrbmYpQCYwq1qbWcC1geVvA++HY/xcRERqV+eQS2BMfBwwB0gCXnDOFZjZw0COc24W8DzwopkVAjuoLPoiIhJBQY2hO+dmA7OrrZtUZbkEGBPaaCIiciw0U1REJE6ooIuIxAkVdBGROKGCLiISJ8zXpwvNrBhYe5x3b0O1WahRRvnqR/nqL9ozKt/x6+qcS6tpg7eCXh9mluOcy/CdozbKVz/KV3/RnlH5wkNDLiIicUIFXUQkTsRqQX/Wd4A6KF/9KF/9RXtG5QuDmBxDFxGRr4vVHrqIiFSjgi4iEieiuqBH88Wpzayzmc01s8VmVmBmE2poc56Z7TazBYGfSTU9VhgzrjGzhYF959Sw3cxscuD45ZvZkAhm61PluCwwsz1mNrFam4gfPzN7wcy2mtmiKutam9k7ZrYi8LtVLfe9NtBmhZldW1ObMGR7wsyWBv5+r5pZy1rue9RzIcwZHzKzDVX+jpfWct+jPt/DmG9qlWxrzGxBLfeNyDGsF+dcVP5Q+VW9K4EeQAqQB/Sr1uYnwJ8Cy5nA1Ajmaw8MCSw3A5bXkO884J8ej+EaoM1Rtl8KvAUYMAz4zOPfejOVEya8Hj/gHGAIsKjKuseB+wLL9wGP1XC/1sCqwO9WgeVWEcg2AmgYWH6spmzBnAthzvgQcFcQ58BRn+/hyldt+5PAJJ/HsD4/0dxD/+/FqZ1zh4EjF6euajTw98DyDOBCM4vIhR6dc5ucc/MDy3uBJVReWzWWjAb+4SrNA1qaWXsPOS4EVjrnjnfmcMg45/5D5Xf6V1X1PPs7cEUNd70EeMc5t8M5txN4BxgZ7mzOubedc2WBm/OovKKYN7Ucv2AE83yvt6PlC9SOa4CsUO83UqK5oNd0cerqBfMrF6cGjlycOqICQz3pwGc1bB5uZnlm9paZ9Y9oMHDA22b2ZeAC3dUFc4wjIZPan0Q+j98R7ZxzmwLLm4F2NbSJhmN5PZX/cdWkrnMh3MYFhoVeqGXIKhqO3zeALc65FbVs930M6xTNBT0mmFlTYCYw0Tm3p9rm+VQOIwwCpgCvRTje2c65IcAo4H/M7JwI779OgcsaXg5Mr2Gz7+P3Na7yf++o+6yvmd0PlAEv1dLE57nwR+BkYDCwicphjWg0lqP3zqP++RTNBT3qL05tZslUFvOXnHOvVN/unNvjnNsXWJ4NJJtZm0jlc85tCPzeCrxK5b+1VQVzjMNtFDDfObel+gbfx6+KLUeGogK/t9bQxtuxNLPrgMuA7wVecL4miHMhbJxzW5xz5c65CuC5Wvbt9VwM1I+rgKm1tfF5DIMVzQU9qi9OHRhvex5Y4px7qpY2Jx0Z0zezoVQe74i84JhZEzNrdmSZyjfPFlVrNgv4YeDTLsOA3VWGFiKl1l6Rz+NXTdXz7Frg9RrazAFGmFmrwJDCiMC6sDKzkcA9wOXOuQO1tAnmXAhnxqrvy1xZy76Deb6H00XAUudcUU0bfR/DoPl+V/ZoP1R+CmM5le9+3x9Y9zCVJy9AKpX/qhcCnwM9IpjtbCr/9c4HFgR+LgVuBW4NtBkHFFD5jv084MwI5usR2G9eIMOR41c1nwHPBI7vQiAjwn/fJlQW6BZV1nk9flS+uGwCSqkcx72Byvdl3gNWAO8CrQNtM4C/VLnv9YFzsRD4UYSyFVI59nzkHDzyqa8OwOyjnQsRPH4vBs6vfCqLdPvqGQO3v/Z8j0S+wPq/HTnvqrT1cgzr86Op/yIicSKah1xEROQYqKCLiMQJFXQRkTihgi4iEidU0EVE4oQKuohInFBBFxGJE/8HGlaJlQznedIAAAAASUVORK5CYII=\n"
          },
          "metadata": {
            "needs_background": "light"
          }
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [],
      "metadata": {
        "id": "KWGJlgl_CuEF"
      },
      "execution_count": null,
      "outputs": []
    }
  ]
}