Titanic_github.ipynb

{
  "nbformat": 4,
  "nbformat_minor": 0,
  "metadata": {
    "colab": {
      "name": "Titanic - github",
      "provenance": [],
      "collapsed_sections": [],
      "toc_visible": true,
      "authorship_tag": "ABX9TyMjDKJ8cdNANxB+4i6QLS8N",
      "include_colab_link": true
    },
    "kernelspec": {
      "name": "python3",
      "display_name": "Python 3"
    }
  },
  "cells": [
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "view-in-github",
        "colab_type": "text"
      },
      "source": [
        "<a href=\"https://colab.research.google.com/github/BrendaLoznik/titanic/blob/main/Titanic_github.ipynb\" target=\"_parent\"><img src=\"https://colab.research.google.com/assets/colab-badge.svg\" alt=\"Open In Colab\"/></a>"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "Ni3RKBcUS7i-"
      },
      "source": [
        "# Welcome to the Titanic!!"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "Ms3uOkWDS3dZ"
      },
      "source": [
        "![header-titanic.jpg](data:image/jpeg;base64,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)"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "aNHF8qUFUeZo"
      },
      "source": [
        "Who hasn’t seen or heard off the world-famous Titanic! The sinking of the Titanic is one of the most famous shipwrecks in history and one of the highest grossing movies of all time. \n",
        "\n",
        "\n",
        "\n",
        "\n",
        "Who would have thought it also the topic of one of the most popular Machine Learning competitions on Kaggle?! Over 19.000 teams and individuals have tried their luck on successfully predicting the fate of the passengers onboard of this unsinkable ship. \n",
        "\n",
        "\n",
        "\n",
        "\n",
        "In this Notebook I will show you how I got a top 9% score on the Leaderboard.\n",
        "\n",
        "\n",
        "\n"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "lKLLmlzpVl5M"
      },
      "source": [
        "![best entry.png](data:image/png;base64,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)"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "Vf3f2ApgluN5"
      },
      "source": [
        "##**1 HOUSEKEEPING**"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "lCAqrb-YZIqN"
      },
      "source": [
        "####**1.1 LOADING LIBRARIES**"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "WWmqBbzblrlN"
      },
      "source": [
        "#basic libraries\n",
        "import numpy as np\n",
        "import pandas as pd\n",
        "import matplotlib.pyplot as plt\n",
        "%matplotlib inline\n",
        "import seaborn as sns\n",
        "\n",
        "#feature engineering\n",
        "import re\n",
        "from sklearn.preprocessing import StandardScaler\n",
        "\n",
        "#modelling\n",
        "from sklearn.model_selection import train_test_split, KFold\n",
        "from sklearn.model_selection import GridSearchCV\n",
        "from sklearn.ensemble import RandomForestClassifier\n",
        "from sklearn.metrics import confusion_matrix, classification_report, roc_auc_score"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "B8gh5S0gZO4i"
      },
      "source": [
        "####**1.2 LOADING DATA**"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "st9satDflvOe",
        "outputId": "0f0e7b1a-b9af-4f05-e7b0-b4746b8451ab",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 129
        }
      },
      "source": [
        "from google.colab import drive \n",
        "drive.mount('/content/gdrive')"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "Go to this URL in a browser: https://accounts.google.com/o/oauth2/auth?client_id=947318989803-6bn6qk8qdgf4n4g3pfee6491hc0brc4i.apps.googleusercontent.com&redirect_uri=urn%3aietf%3awg%3aoauth%3a2.0%3aoob&scope=email%20https%3a%2f%2fwww.googleapis.com%2fauth%2fdocs.test%20https%3a%2f%2fwww.googleapis.com%2fauth%2fdrive%20https%3a%2f%2fwww.googleapis.com%2fauth%2fdrive.photos.readonly%20https%3a%2f%2fwww.googleapis.com%2fauth%2fpeopleapi.readonly&response_type=code\n",
            "\n",
            "Enter your authorization code:\n",
            "··········\n",
            "Mounted at /content/gdrive\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "ujW4z6_vl0te"
      },
      "source": [
        "train=pd.read_csv('/content/gdrive/My Drive/Kaggle/Titanic/train.csv')\n",
        "test=pd.read_csv('/content/gdrive/My Drive/Kaggle/Titanic/test.csv')"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "sYTRTGCBSSE5"
      },
      "source": [
        "##**2 CLEANING**"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "-XBlLlB1W0Bh"
      },
      "source": [
        "I like to concatenate the training and testing data together for cleaning and exploratory data analysis. We have to be careful in order to prevent data leakage."
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "LqHTKbjDVw7t",
        "outputId": "9f68693c-6e37-42b5-ccbe-c22a1bce5fc2",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 206
        }
      },
      "source": [
        "# Concatenate training and testing data together.\n",
        "raw = pd.concat([train, test])\n",
        "raw.head()"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/html": [
              "<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>PassengerId</th>\n",
              "      <th>Survived</th>\n",
              "      <th>Pclass</th>\n",
              "      <th>Name</th>\n",
              "      <th>Sex</th>\n",
              "      <th>Age</th>\n",
              "      <th>SibSp</th>\n",
              "      <th>Parch</th>\n",
              "      <th>Ticket</th>\n",
              "      <th>Fare</th>\n",
              "      <th>Cabin</th>\n",
              "      <th>Embarked</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>0</th>\n",
              "      <td>1</td>\n",
              "      <td>0.0</td>\n",
              "      <td>3</td>\n",
              "      <td>Braund, Mr. Owen Harris</td>\n",
              "      <td>male</td>\n",
              "      <td>22.0</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>A/5 21171</td>\n",
              "      <td>7.2500</td>\n",
              "      <td>NaN</td>\n",
              "      <td>S</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1</th>\n",
              "      <td>2</td>\n",
              "      <td>1.0</td>\n",
              "      <td>1</td>\n",
              "      <td>Cumings, Mrs. John Bradley (Florence Briggs Th...</td>\n",
              "      <td>female</td>\n",
              "      <td>38.0</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>PC 17599</td>\n",
              "      <td>71.2833</td>\n",
              "      <td>C85</td>\n",
              "      <td>C</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2</th>\n",
              "      <td>3</td>\n",
              "      <td>1.0</td>\n",
              "      <td>3</td>\n",
              "      <td>Heikkinen, Miss. Laina</td>\n",
              "      <td>female</td>\n",
              "      <td>26.0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>STON/O2. 3101282</td>\n",
              "      <td>7.9250</td>\n",
              "      <td>NaN</td>\n",
              "      <td>S</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>3</th>\n",
              "      <td>4</td>\n",
              "      <td>1.0</td>\n",
              "      <td>1</td>\n",
              "      <td>Futrelle, Mrs. Jacques Heath (Lily May Peel)</td>\n",
              "      <td>female</td>\n",
              "      <td>35.0</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>113803</td>\n",
              "      <td>53.1000</td>\n",
              "      <td>C123</td>\n",
              "      <td>S</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>4</th>\n",
              "      <td>5</td>\n",
              "      <td>0.0</td>\n",
              "      <td>3</td>\n",
              "      <td>Allen, Mr. William Henry</td>\n",
              "      <td>male</td>\n",
              "      <td>35.0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>373450</td>\n",
              "      <td>8.0500</td>\n",
              "      <td>NaN</td>\n",
              "      <td>S</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>"
            ],
            "text/plain": [
              "   PassengerId  Survived  Pclass  ...     Fare Cabin  Embarked\n",
              "0            1       0.0       3  ...   7.2500   NaN         S\n",
              "1            2       1.0       1  ...  71.2833   C85         C\n",
              "2            3       1.0       3  ...   7.9250   NaN         S\n",
              "3            4       1.0       1  ...  53.1000  C123         S\n",
              "4            5       0.0       3  ...   8.0500   NaN         S\n",
              "\n",
              "[5 rows x 12 columns]"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 10
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "xQ54Kaf_Vw_L",
        "outputId": "cba74ca7-112f-461d-d893-736a94b5616e",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 368
        }
      },
      "source": [
        "# The dataset consists of a mixture of numerical and non-numerical features\n",
        "raw.info()"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "<class 'pandas.core.frame.DataFrame'>\n",
            "Int64Index: 1309 entries, 0 to 417\n",
            "Data columns (total 12 columns):\n",
            " #   Column       Non-Null Count  Dtype  \n",
            "---  ------       --------------  -----  \n",
            " 0   PassengerId  1309 non-null   int64  \n",
            " 1   Survived     891 non-null    float64\n",
            " 2   Pclass       1309 non-null   int64  \n",
            " 3   Name         1309 non-null   object \n",
            " 4   Sex          1309 non-null   object \n",
            " 5   Age          1046 non-null   float64\n",
            " 6   SibSp        1309 non-null   int64  \n",
            " 7   Parch        1309 non-null   int64  \n",
            " 8   Ticket       1309 non-null   object \n",
            " 9   Fare         1308 non-null   float64\n",
            " 10  Cabin        295 non-null    object \n",
            " 11  Embarked     1307 non-null   object \n",
            "dtypes: float64(3), int64(4), object(5)\n",
            "memory usage: 132.9+ KB\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "K95WAQ94Xs_Z"
      },
      "source": [
        "The dataset has 1.309 rows and 12 different features.\n",
        "I will be easier to clean and analyze these features when I first divide them in different categories.\n",
        "\n",
        "* Nominal: Categorical features without order\n",
        "* Ordinal: Categorical features with order\n",
        "* Binary: Categorical features with two possible outcomes\n",
        "* Discrete: Numerical features that take discrete values\n",
        "* Continuous: Numerical features consisting of continuous values"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "ZaV_zv-cVxCB"
      },
      "source": [
        "#define feature categories\n",
        "nominal = ['Name', 'Ticket', 'Embarked']\n",
        "ordinal =  ['Cabin']\n",
        "binary = ['Survived', 'Sex']\n",
        "discrete = ['PassengerId', 'Pclass', 'SibSp', 'Parch' ]\n",
        "continious = ['Age', 'Fare' ]"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "QJ1aERIqVxGl",
        "outputId": "db8f5341-316f-47bd-e3fa-e98c46b65213",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 128
        }
      },
      "source": [
        "print(\"Number of nominal features: {}\".format(len(nominal)))\n",
        "print(\"Number of ordinal features: {}\".format(len(ordinal)))\n",
        "print(\"Number of discrete features: {}\".format(len(discrete)))\n",
        "print(\"Number of continious features: {}\".format(len(continious)))\n",
        "print(\"Number of binary features: {}\".format(len(binary)))\n",
        "print(\"Total number offeatures: {}\".format(len(continious + discrete + ordinal + nominal + binary)))"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "Number of nominal features: 3\n",
            "Number of ordinal features: 1\n",
            "Number of discrete features: 4\n",
            "Number of continious features: 2\n",
            "Number of binary features: 2\n",
            "Total number offeatures: 12\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "IXwLzs-4YjfR"
      },
      "source": [
        "I generally like to make a copy of the dataset for each phase of the project. This makes it easier to make changes."
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "apr3-6JUVxKG"
      },
      "source": [
        "cleaned = raw.copy()"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "plciiBCZY4CK"
      },
      "source": [
        "####**2.1 CONTINUOUS  FEATURES**"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "PbmP3eEVVxEx",
        "outputId": "db723c5a-4917-4632-e19e-a361f87eab0e",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 72
        }
      },
      "source": [
        "# number of missing continious data\n",
        "print(raw[continious].isna().sum().sort_values(ascending=False))"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "Age     263\n",
            "Fare      1\n",
            "dtype: int64\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "9SLhCbQGZYnC"
      },
      "source": [
        "**2.1.1 AGE**"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "PkJcfx1ZZgV9"
      },
      "source": [
        "We don't know the age of 263 passengers. Let's figure out the best way to impute these missing ages."
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "PRtm0SrmZVoq",
        "outputId": "eb4aebcb-216b-4cf3-9e55-1c34ddcdfd7e",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 295
        }
      },
      "source": [
        "# Older people travel in higher classes: Class and Gender are a factor\n",
        "plt.figure(figsize=(6,4))\n",
        "ax =sns.boxplot(x= \"Pclass\", y = \"Age\", data = raw, hue = \"Sex\", palette = 'Set2')\n",
        "ax.set(title = 'Age by Class and Gender')\n",
        "plt.legend(loc= \"upper right\")\n",
        "plt.show()"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<Figure size 432x288 with 1 Axes>"
            ]
          },
          "metadata": {
            "tags": [],
            "needs_background": "light"
          }
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "aTwOIl-qZVyN"
      },
      "source": [
        "#extract title with regex\n",
        "cleaned[\"Title\"] = cleaned[\"Name\"].apply(lambda x: re.search('([A-Z][a-z]+)\\.', x).group(1))"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "9tUPZwRVZV4F",
        "outputId": "01ce3bae-5df3-4cf2-a903-5efda59c6746",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 295
        }
      },
      "source": [
        "# Ages differ quite drastically within genders\n",
        "plt.figure(figsize=(15,4))\n",
        "ax =sns.boxplot(x= \"Title\", y = \"Age\", data = cleaned, palette = 'Set2')\n",
        "ax.set(title = 'Age by Title')\n",
        "plt.show()"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "image/png": "iVBORw0KGgoAAAANSUhEUgAAA3UAAAEWCAYAAADB8b79AAAABHNCSVQICAgIfAhkiAAAAAlwSFlzAAALEgAACxIB0t1+/AAAADh0RVh0U29mdHdhcmUAbWF0cGxvdGxpYiB2ZXJzaW9uMy4yLjIsIGh0dHA6Ly9tYXRwbG90bGliLm9yZy+WH4yJAAAgAElEQVR4nO3de5xddX3v/9cnGYjhmuEig0EnHC9QtUjrVPSMTlDUo79a7aTVYm0DP+yPn49y0V5+x0sPqUJtobanPw1oDwqWWC94yQi1rdWiSewcoQbRiGBEJKMJDASYAIkxZpLP+WOtCROYy05m7732nnk9H495zF5rr1nrvffsvfb+rO9a329kJpIkSZKk9jSv6gCSJEmSpINnUSdJkiRJbcyiTpIkSZLamEWdJEmSJLUxizpJkiRJamMWdZIkSZLUxizqJEkCImJNRPxBg7fxjIjYHhHzp1gmI+JZjcwhSZpdLOokSS2vLLhGImJB1VmmEhEvK4u27RGxoyzQxqa3A2TmEZm5p1y+4YWkJGn2s6iTJLW0iFgCvAxI4PWVhplGZn6jLNqOAJ5Xzl40Ni8zf1JlPknS7GRRJ0lqdcuBm4F/AM4Zf0dEHBsR/xQRj0bEtyLiLyLiP8bdf2pEfDUiHo6IjRHxpmm29cyI+M9yfTdExDHlev45Ii56wrY3RET/gTyQiFhStt51RMT7KYrVK8uWvCsnWH5BRPxNRPwkIu6PiL+PiIUHsk1J0uxnUSdJanXLgU+WP/8tIk4Yd99VwA6gi6Lg21f0RcThwFeBTwFPBc4GPhwRz51mW+cBJwKjwIfK+dcBvzdu3S8AFgP/fLAPKjP/DPgGcGHZinfhBItdDjwHOB14VrnNFQe7TUnS7GRRJ0lqWRHxUqAb+Gxm3grcDfxued984LeAP8/Mn2XmHRTF15jXAZsy8+OZOZqZtwFfAN44xSY/kZm3Z+YO4BLgTeV2bgSeExHPLpf7feD6zPxF/R7t/iIigPOBP8rMhzPzMeAvKYpTSZL2saiTJLWyc4CvZOaD5fSneLw17nigA/jpuOXH3+4GzoiIbWM/wFsoWvUmM/7vh4BDgOMy8+fA9cDvRcQ84M3AJw7yMdXqeOAw4NZx+b9czpckaZ+OqgNIkjSR8tqxNwHzI2K4nL0AWFSe/ng7xSmSJwE/LO9/+rhV/BRYm5mvOoDNjv/7ZwC7gbGC8jqKQu4/gJ9l5jcPYL2TySnuexDYCTwvM7fUYVuSpFnKljpJUqv6TWAP8FyKa8pOB36J4jq05eWwAKuB90bEYRFxKsU1cWO+RHHK5O9HxCHlz69FxC9Nsc3fi4jnRsRhwKXA58eGHyiLuL3A31K/Vrr7gf8y0R2ZuRf4KPB3EfFUgIhYHBH/rU7bliTNEhZ1kqRWdQ7w8cz8SWYOj/0AVwJviYgO4ELgaGCYotD6NLALoLwG7dUU16DdWy5zBUVr32Q+QdHL5jDwFODiJ9y/Cvhl4B/r8QCBDwK/XY7B96EJ7n8n8CPg5oh4FPh34JQ6bVuSNEtE5lRnfkiS1D4i4gqgKzPPmXbhg1v/cuD8zHxpI9YvSdLBsKVOktS2ynHoTovCi4C3AgMN2tZhwB8CVzdi/ZIkHSyLOklSOzuS4rq6HRS9U/4tcEO9N1Jex7aV4hq4T9V7/ZIkzYSnX0qSJElSG7OlTpIkSZLaWFuMU3fcccflkiVLqo4hSZIkSZW49dZbH8zM4ye6ry2KuiVLlrB+/fqqY0iSJElSJSJiaLL7PP1SkiRJktqYRZ0kSZIktTGLOkmSJElqYxZ1kiRJktTG5nRRNzIywqWXXsq2bduqjiJJkiRJB6WhRV1E/FFEfD8ibo+IT0fEUyLi5Ii4JSJ+FBHXR8ShjcwwlYGBATZu3Mjq1auriiBJkiRJM9Kwoi4iFgMXAz2Z+XxgPnA2cAXwd5n5LGAEeGujMkxlZGSEtWvXkpmsW7fO1jpJkiRJbanRp192AAsjogM4DLgPeAXw+fL+64DfbHCGCQ0MDJCZAOzdu9fWOkmSJEltqWFFXWZuAf4G+AlFMfcIcCuwLTNHy8U2A4sn+vuIOD8i1kfE+q1bt9Y93+DgIKOjRYzR0VEGBwfrvg1JkiRJarRGnn7ZCbwBOBl4GnA48Jpa/z4zr87MnszsOf744+uer7e3l46ODgA6Ojro7e2t+zYkSZIkqdEaefrlK4F7MnNrZu4GVgO9wKLydEyAk4AtDcwwqf7+fiICgHnz5rFs2bIqYkiSJEnSjDSyqPsJ8OKIOCyK6uks4A7g68Bvl8ucA9zQwAyT6uzsZOnSpUQEfX19LFq0qIoYkiRJkjQjjbym7haKDlG+DXyv3NbVwDuBP46IHwHHAtc0KsN0+vv7OeWUU2ylkyRJktS2YqwHyFbW09OT69evrzqGJEmSJFUiIm7NzJ6J7mv0kAZqkJGRES699FLH15MkSZLmOIu6NjUwMMDGjRsdX0+SJEma4yzq2tDIyAhr164lM1m3bp2tdZIkSdIcZlHXhgYGBhi7FnLv3r221kmSJElzmEVdGxocHGR0dBSA0dFRBgcHK04kSZIkqSpzuqhr185Gent76egoxm/v6Oigt7e34kSSJEmSqjKni7p27Wykv7+fYjx3mDdvnuPsSZIkSXPYnC3q2rmzkc7OTpYuXUpE0NfXx6JFi6qOJEmSJKkic7aoa/fORvr7+znllFNspZMkSZLmuDlb1LV7ZyOdnZ2sWLHCVjpJkiRpjpuzRZ2djUiSJEmaDeZsUWdnI5IkSZJmgzlb1NnZiCRJkqTZYM4WdQA9PT0AvOhFL6o4iSRJkiQdnDld1F133XVkJh//+MerjiJJkiRJB2XOFnWbNm3ivvvuA+C+++5jaGio4kSSJEmSdOAaVtRFxCkR8Z1xP49GxDsi4piI+GpE3FX+7mxUhqmsXLlyv+kPfehDVcSQJEmSpBlpWFGXmRsz8/TMPB14IfAzYAB4F3BTZj4buKmcbrqxVrrJpiVJkiSpHXQ0aTtnAXdn5lBEvAE4s5x/HbAGeGeTckiSJGkWW7VqFevWrZt2uZ07d5KZddlmRLBw4cJpl+vr62P58uV12aY0XrOuqTsb+HR5+4TMHGsWGwZOmOgPIuL8iFgfEeu3bt1a90Dz58+fclqSJEmS2kHU6wjFpBuIOBS4F3heZt4fEdsyc9G4+0cyc8rr6np6enL9+vV1zfWRj3yEb3zjG/um+/r6eNvb3lbXbUiSJElSPUTErZnZM9F9zWipey3w7cy8v5y+PyJOLIOdCDzQhAxPcvbZZxMRlDk4++yzq4ghSZIkSTPSjKLuzTx+6iXAjcA55e1zgBuakOFJOjs79w06fsYZZ7Bo0aJp/kKSJEmSWk9Di7qIOBx4FbB63OzLgVdFxF3AK8vpShx66KH7/ZYkSZKkdtPQoi4zd2TmsZn5yLh5D2XmWZn57Mx8ZWY+3MgMkxkZGeHmm28G4Oabb2bbtm1VxJAkSZKkGWlW75ctZ2BggL179wKwZ88eVq9ePc1fSJIkSVLrmbNF3eDgIHv27AGKom5wcLDiRJIkSZJ04OZsUdfTs39voL/2a79WURJJkiRJOnhztqh7okaP1ydJkiRJjTBni7pbbrllymlJkiRJagdztqgb6yRlsmlJkiRJagdztqgbHR2dclqSJEmS2sGcLeoWLFgw5bQaY2RkhEsvvdRxASVJkqQ66ag6gPa3atUqhoaGpl1ueHgYgK6urimX6+7uZvny5XXJVg8DAwNs3LiR1atXc95551UdR5IkSWp7c7albteuXVNOt7pdu3a1XeaRkRHWrl1LZrJu3Tpb6yRJkqQ6mLMtdR0dHftdR9fR0RpPRa2tapdddhkAl1xySSPj1NXAwMC+oSP27t1ra50kSZJUB3O2pW7evHlTTqv+BgcH9xXSo6OjDA4OVpxIkiRJan9ztpJ5znOes9/0KaecUlGSuaO3t3dfi2hHRwe9vb0VJ5IkSZLa35wt6u688879pu+4446Kkswd/f39RARQtIwuW7as4kSSJElS+5uzRd2ePXumnFb9dXZ2snTpUiKCvr4+Fi1aVHUkSZIkqe21Ru8gmjP6+/vZvHmzrXSSJElSnTS0pS4iFkXE5yPiBxFxZ0S8JCKOiYivRsRd5e/ORmZQa+ns7GTFihW20kmSJEl10ujTLz8IfDkzTwVeANwJvAu4KTOfDdxUTkuSJEmSDkLDTr+MiKOBPuBcgMz8BfCLiHgDcGa52HXAGuCdjcohSZKkxli1ahVDQ0PTLjc8PAxAV1fXlMt1d3fXPGavpMc1sqXuZGAr8PGIuC0iPhYRhwMnZOZ95TLDwAkT/XFEnB8R6yNi/datWxsYU5IkSY20a9cudu3aVXUMadZqZEcpHcCvAhdl5i0R8UGecKplZmZE5ER/nJlXA1cD9PT0TLjMVKY7cnTIIYewe/fu/aYvu+yyCZf1qFH9bNq0icsuu4wVK1bQ3d1ddRxJkjQDtX4/GvuOdckllzQyjjRnNbKlbjOwOTNvKac/T1Hk3R8RJwKUvx9oYIZJLV68eL/pk046qYoYc85VV13Fzp07ufLKK6uOIkmSJM0KDWupy8zhiPhpRJySmRuBs4A7yp9zgMvL3zc0Yvu1HDk655xz2L17N11dXbz//e9vRAyNs2nTJrZs2QLAli1bGBoasrVOkiRJmqFG9355EfDJiNgAnA78JUUx96qIuAt4ZTldicWLFxMRvP3tb68qwpxy1VVX7Tdta50kSZI0cw0dfDwzvwP0THDXWY3cbq0WLlzIqaeeamtRk4y10k02LUmSJOnANbSok8ZbvHjxfoXcE69r1NxhF9iS1Ppq3VfXYmw9k3VKdyDc50tPZlGnprngggt4z3ves2/6wgsvrDCN2oHdX0tSdYaGhrjn7jt42vF7Zryujiiu+Nn16PdmtJ57t86fcRZpNrKoU9MsWbJkX2vd4sWLPe11DrMLbElqD087fg9/+MbtVcfY58OfO6LqCFJLanRHKdJ+LrjgAhYuXGgrnSRJklQnttSpqZYsWcI111xTdQxJkiRp1rClTk01MjLCpZdeyrZt26qOIkmSJM0KttSpqQYGBti4cSOrV6/mvPPOqzqOJEmaxPDwMDt3zG+p69ju3TqfhT8brjqG1HJsqVPTjIyMsHbtWjKTdevW2VonSZIk1YEtdWqagYEBMhOAvXv32lqntlTLuE2OrydpNujq6mLXo1tbrvfLBUdNvW+V5iJb6tQ0g4ODjI6OAjA6Osrg4GDFiaTG2LVrl2PsSZKkprGlTk3T29vLmjVrGB0dpaOjg97e3qojSQeslpY1x9eTJEnNZEudmqa/v5+IAGDevHksW7as4kSSJElS+7OoU9N0dnaydOlSIoK+vj4WLVpUdSRJkiSp7Xn6pZqqv7+fzZs320o3i9XSkUitxtYzdjrjTNgpiZqt1veCHetIkmbKok5N1dnZyYoVK6qOoQYaGhrinrvv4GnH75nxujqiOJlg16Pfm9F67t06f8ZZpEaxUx21snu31mecuge3Ffvz4xbtnXGek4+acRxp1mloURcRm4DHgD3AaGb2RMQxwPXAEmAT8KbMHGlkDrWODRs2cMUVV/Dud7+b5z//+VXHUYM87fg9LdcFttRstbaq2bGOWlV3d3fd1jU6UrRaLzhqZus8+aj65pJmi2a01L08Mx8cN/0u4KbMvDwi3lVOv7MJOdQCVq5cSWbywQ9+kI9+9KNVx5H2adXTRsHT7iRVo577HQ9eSI1VxemXbwDOLG9fB6zBom5O2LBhAzt27ABgx44d3H777bbWqWUUp43eRddxR854XfOjOPV05yPDM17X8IOPzXgdqr9WPQjgAQBJmpsaXdQl8JWISOB/ZebVwAmZeV95/zBwQoMzqEWsXLlyv2lb69Rquo47kvPe0FN1jP1ce8P6qiNoAkNDQ9zzwx+w+IhDZryuQ0ZHAfjFvXfPaD1btu+ecRZJUntqdFH30szcEhFPBb4aET8Yf2dmZlnwPUlEnA+cD/CMZzyjwTHVDGOtdJNNS1I7WXzEIVx02lOrjrHPyg0PVB1BklSRhhZ1mbml/P1ARAwALwLuj4gTM/O+iDgRmPBTqGzVuxqgp6dnwsJP7eXwww/fr5A7/PDDK0wjSWonq1atYt26dVMus3PnTjLr95UhIli4cOGUy/T19XnKq6TKNayoi4jDgXmZ+Vh5+9XApcCNwDnA5eXvGxqVQa3loosu4vLLL983/fa3v73CNJJ08IaHh9m5fXdLtY5t2b6bhcMzv45TktR+GtlSdwIwEBFj2/lUZn45Ir4FfDYi3goMAW9qYAa1kNNOO21fa93hhx9uJymS1GS1tHZBfVu8amntgulbvJYvX26LmCRNomFFXWb+GHjBBPMfAs5q1HbV2i666CKuuOIKW+nUcoaHh/nZjsdarmOS+x58jMN2Vp1CT9TV1cUv9u5ouWvqDu3qqjqGJKkCVQxpoDnstNNO45Of/GTVMSRpTrK1S5JmJ4s6SaJoedn5CC05pMHCo219kdSaah2zsdbxGB1rUTo4FnWSJB2ELXXqKOXBncU4dcctnNlH8pbtuzl5xmmkxliwYEHVEaRZzaJOTbVp0yYuu+wyVqxYQXd3d9VxJOmg1HP/tbtswTj0aTNb58nUN5dUC1vVpNZgUaemuuqqq9i5cydXXnklH/jAB6qOI0kHpZ5fZMdOR7vkkkvqtk5J0twyr+oAmjs2bdrEli1bANiyZUtN5+BLkiRJmpotdU1U68XEtaj1guNaNOui5Kuuumq/aVvrJM1mdiAhSWoWi7omGhoa4p4f/oDFRxwy43UdMlpcWP+Le++e0Xq2bN894yw1b6tspZtsWpLmIjuQkCTNlEVdky0+4pCWG6y2WRYvXrxfIbd48eKmbVuSms1WNUlSs3hNnZrmggsu2G/6wgsvrCiJJEmSNHtY1KlplixZsq91bvHixXa9LUmSJNWBRZ2a6oILLmDhwoW20kmSJEl14jV1aqolS5ZwzTXXVB1DkiRJmjWmLeoi4gTgL4GnZeZrI+K5wEsy02/m2k8t3XcPDw8D0NXVNeVydt0tSZIk1aaW0y//Afg34Gnl9A+BdzQqkGa3Xbt2sWvXrqpjSJIkSbNGLadfHpeZn42IdwNk5mhE7GlwLrWhWlrWxgbXveSSSxodp+5GRkZYuXIlF198MYsWLao6TssaHh5m5475fPhzR1QdZZ97t85n4c+Gq44hSZLUELUUdTsi4lggASLixcAjtW4gIuYD64Etmfm6iDgZ+AxwLHAr8PuZ+YsDTi412cDAABs3bmT16tWcd955VcdRAww/+BjX3rB+xut56JGfAXDs0YfNeF3DDz7GyUdPfbqyJEma22op6v4YuBF4ZkQMAscDv30A23g7cCdwVDl9BfB3mfmZiPh74K3ARw5gfVLTjYyMsHbtWjKTdevWsWzZMlvrJtHV1cWuR7fyh2/cXnWUfT78uSNYcNT013HWy55txbWlC+tQjJ18dJfDf0iSpClNW9Rl5rcjYilwChDAxszcXcvKI+Ik4NeB9wN/HBEBvAL43XKR64D3YlGnFjcwMEBmArB3796Gt9atWrWKdevWTbvczp079+WaqYhg4cKF0y7X19c3KzuxqedjaufTjCVJUvuZtqOUiFgGvJ6iqHsO8BsRcVZEPLWG9f//wH8H9pbTxwLbMnO0nN4MLJ5ku+dHxPqIWL9169YaNiU1zuDgIKOjxct2dHSUwcHBihNJkiRJhVpOv3wr8BLg6+X0mRTXwp0cEZdm5icm+qOIeB3wQGbeGhFnHmiwzLwauBqgp6enPk0R0kHq7e1lzZo1jI6O0tHRQW9vb0O3t3z58lnZGiZJkqT6q6Wo6wB+KTPvh33j1q0CzgDWARMWdUAv8PqI+L+Ap1BcU/dBYFFEdJStdScBW2b2EKTG6+/vZ+3atQDMmzePZcuWVZxIkiRJM1XLJS/1vNwFarvk5UAvd6llnLqnjxV0pQfKeQ8Dk15bl5nvzsyTMnMJcDbwtcx8C0WL31hHK+cAN9ScVqpIZ2cnS5cuJSLo6+uzkxRJkiS1jFpa6tZExJeAz5XTv1XOOxzYdhDbfCfwmYj4C+A24JoD+eNVq1YxNDR0EJt9srH1jHVqMFPd3d2eMjeL9ff3s3nzZlvpJEmSxmnnDt5myyUvtRR1FwDLgJeW0+uBEzJzB/DyWjaSmWuANeXtHwMvOtCgY4aGhvjBj+7ikGOOPNhV7DNKMYb63Q/PfFDi3Q8/NuN1qLV1dnayYsWKqmNIkiRJ+6llSIOMiB8DLwbeCNwDfKHRwaZyyDFHcuyrz6gywpM89JVbqo4gSZIkNd1sae1qZ5MWdRHxHODN5c+DwPVAZGZNrXOSJEmSpMabqqXuB8A3gNdl5o8AIuKPmpJKkiRJklSTqXq/XAbcB3w9Ij4aEWcB0ZxYkiRJkqRaTFrUZeYXM/Ns4FSKYQjeATw1Ij4SEa9uVkBJkiRJ0uSmHacuM3dk5qcy8zcoBgu/jWJYAkmSJElSxWoZ0mCfzBwBri5/dICGh4fZuX03Kzc8UHWUfbZs383C4ZkP6SBJkiSpGtO21EmSJEmSWtcBtdRpZrq6uvjF3h1cdNpTq46yz8oND3BoV1fVMSRJkiQdJFvqJEmSJKmNWdRJkiRJUhvz9EtJkiSpBaxatYp169ZNuczOnTvJzLptMyJYuHDhlMv09fWxfPnyum1T9WdLnSRJkiS1MVvqpBqNjIywcuVKLr74YhYtWlR1HFVk1apVDA0NTbnM2P2XXXbZlMt1d3d75FOStM/y5cv9XNBBsaVOqtHAwAAbN25k9erVVUdRi1uwYAELFiyoOoYkSZojbKmTajAyMsLatWvJTNatW8eyZctsrZujPIIqSZJaTcNa6iLiKRHxnxHx3Yj4fkS8r5x/ckTcEhE/iojrI+LQRmWQ6mVgYGDfRcl79+61tU6SJEkto5GnX+4CXpGZLwBOB14TES8GrgD+LjOfBYwAb21gBqkuBgcHGR0dBWB0dJTBwcGKE0mSJEmFhp1+mUWzxvZy8pDyJ4FXAL9bzr8OeC/wkUblkOqht7eXNWvWMDo6SkdHB729vVVHkuaMWjqnGR4eBqCrq2vK5eycRpI0GzX0mrqImA/cCjwLuAq4G9iWmaPlIpuBxZP87fnA+QDPeMYzGhlTmlZ/fz9r164FYN68eSxbtqziRK3t3q3z+fDnjpjxeh7cVpxMcNyivTPOc/JRM46jOqulWIOiYNu1a9eUy/z85z8HmHa54eHhmrZp8SdJaicNLeoycw9wekQsAgaAUw/gb68Grgbo6emp3wiL0kHo7Oxk6dKl3HTTTfT19dlJyhS6u7vrtq7RkeLL94KjZrbOk4+qby7Vx9DQED/40V0ccsyRUy94KHDodL2J7gFg9MiplxsF7n54eMpldj/82DTbkiSptTSl98vM3BYRXwdeAiyKiI6yte4kYEszMkgz1d/fz+bNm22lm0Y9WzfGxnm75JJL6rZOtY6xUybroePIw+q2LqhvNkmSGq1hRV1EHA/sLgu6hcCrKDpJ+Trw28BngHOAGxqVQaqnzs5OVqxYUXUMaVbJ0dG6tIzlnqKlLubPn/m6RkenX0iSpBbSyJa6E4Hryuvq5gGfzcwvRcQdwGci4i+A24BrDmSlw8PD7N7+GA995Zb6J56B3Q8/xvAvqk7RGLVe91KLsfWMtcDMlNe9SO3rjDPOqN81dXuKa+qecsjUo+QsWLBg2s5UwNN1JUntpZG9X24AfmWC+T8GXtSo7ar+hoaG+OFdd3P4UcfNeF2jewKALfc/MuN17Xj0wRmvQ1J1aj0gY++XkiRNrSnX1NVTV1cXOx6GY199RtVR9vPQV26h65jpj/5u2b6blRsemPH2HtxZnB503MKZ/Qu3bN/NyTUsd/hRx3Hai39zRtuqtw03f7Gp2xsZGWHlypVcfPHFdpQyQ7W2/tbasusX9dnN/60kSVNru6KundXzdJ7d5ZfdQ582w14B8TSjWg0MDLBx40ZWr17NeeedV3WcOWHBgul6PJQkSZJFXRPZK2D7GhkZYe3atWQm69atY9myZbbWzYAtL5IkSfUzr+oAUjsYGBggsxguce/evaxevbriRJIkSVLBok6qweDgIKNlN+ejo6MMDg5WnEiSJEkqWNRJNejt7WV+Of7V/Pnz6e3trTiRJEmSVLCok2rQ39/P3r17AchMli1bVnEiSZIkqWBRJ9UoIqqOoDYxMjLCpZdeyrZt26qOIkmS5oC27P1y98OP8dBXbpnxekYf+xkAHUceNuN17X74MahhnDq1p4GBgX1FXUQ4rIGm1OzhL1atWsW6deumXGbnzp37Ovuph4hg4cKFUy7T19dnT6ea82p5f0J936O1vD/B96g0m7RdUVfPMdWGHivGeuuuRzF2TJfjvc1ig4OD7NmzB4A9e/YwODhoUacJOfyFJElqtrYr6hzrTVXo7e1lzZo1jI6O0tHRYUcpmtREw180+gDA8uXLPdoutSjfn5KawWvqpBr09/fvO/1y3rx5dpSiSTn8hSRJajaLOqkGnZ2dLF26lIigr6/P0+k0qd7eXjo6ipMgbNWVJEnNYFEn1ai/v59TTjnFVjpNyVZdSZLUbBZ1Uo06OztZsWKFrXSakq26kiSp2dquoxQ13/DwMDse28GGm79YdZT97Hj0QYZzZ9UxpCfp7+9n8+bNttJJkqSmaFhLXUQ8PSK+HhF3RMT3I+Lt5fxjIuKrEXFX+buzURkkqQq26kqSpGZqZEvdKPAnmfntiDgSuDUivgqcC9yUmZdHxLuAdwHvbGAOzVBXVxd74hFOe/FvVh1lPxtu/iJdJxxddQxJkiSpUg1rqcvM+zLz2+Xtx4A7gcXAG4DrysWuA1qrUpAkSZKkNtKUjlIiYgnwK8AtwAmZeV951zBwwiR/c35ErI+I9Vu3bm1GTEmSJElqOw0v6iLiCOALwDsy89Hx92VmAjnR32Xm1ZnZk5k9xx9/fKNjSpIkSVJbamhRFxGHUBR0n8zM1eXs+yPixPL+E4EHGplBkiRJkmazRvZ+GcA1wJ2Z+T/H3XUjcE55+xzghkZlkCRJkqTZrpG9X/YCv4/dDLgAABRZSURBVA98LyK+U857D3A58NmIeCswBLypgRkkSZIkaVZrWFGXmf8BxCR3n9Wo7UqSJEnSXNLIljrNIjsefZANN39xxuv5+Y5HAHjK4TMfX27How+C49RJkiRpjrOo07S6u7vrtq6hoW0ALK5HMXbC0XXNJkmSJLUjizpNa/ny5XVb12WXXQbAJZdcUrd1SpIkSXNZUwYflyRJkiQ1hkWdJEmSJLUxizpJkiRJamNeU6c5b9WqVQwNDU273PDwMABdXV1TLtfd3V3X6xAlSZKkqVjUSTXatWtX1REkSZKkJ7Go05xXa6uaPXdKkiSpFXlNnSRJkiS1MYs6SZIkSWpjFnWSJEmS1MYs6iRJkiSpjVnUSZIkSVIbs6iTJEmSpDbWsCENIuJa4HXAA5n5/HLeMcD1wBJgE/CmzBxpxPZrGVB67P6xruon42DSkiRJklpVI1vq/gF4zRPmvQu4KTOfDdxUTldmwYIFLFiwoMoIkiRJkjQjDWupy8x1EbHkCbPfAJxZ3r4OWAO8sxHbt2VNkiRJ0lzQ7GvqTsjM+8rbw8AJTd6+JEmSJM0qlXWUkpkJ5GT3R8T5EbE+ItZv3bq1ickkSZIkqX007PTLSdwfESdm5n0RcSLwwGQLZubVwNUAPT09kxZ/0lRq6TCnVrV2rFMLO9+RJElSvTS7qLsROAe4vPx9Q5O3rzlmaGiIe354N11HHDvjdc0fLX7vvHfbjNYzvP2hGWeRJEmSxjRySINPU3SKclxEbAb+nKKY+2xEvBUYAt7UqO1LY7qOOJbzTnt91TH2uXbDjVVHkCRJ0izSyN4v3zzJXWc1apuSJEmSNNdU1lGKJEmSJGnmLOokSZIkqY1Z1EmSJElSG7OokyRJkqQ2ZlEnSZIkSW3Mok6SJEmS2phFnSRJkiS1MYs6SZIkSWpjDRt8XAdn1apVDA0NTbvc2DKXXXbZlMt1d3ezfPnyumRrR8PDw/xs+w6u3XBj1VH2uW/7Qxw2/POqY0iSJGmWsKhrUwsWLKg6giRJkqQWYFHXYuZyq1ojdHV1sXPvNs477fVVR9nn2g03srBrUdUxJEmSNEt4TZ0kSZIktTGLOkmSJElqY3P69MuRkRFWrlzJxRdfzKJFng43Ww1vf6guHaU8tPMRAI5dePSM85yMrzdJkiTVx5wu6gYGBti4cSOrV6/mvPPOqzqOGqC7u7tu69ozVBR1C582s4LsZBbVNZckSZLmtjlb1I2MjLB27Voyk3Xr1rFs2TJb65pg9+7dbNmyhW3btjXl+a5nxzNjw0dccskldVunJEmSNFOVFHUR8Rrgg8B84GOZeXmzMwwMDJCZAOzdu9fWujqoZYy9e+65hz179vCe97yHE088cdLl5vr4epIkSVKtmt5RSkTMB64CXgs8F3hzRDy32TkGBwcZHR0FYHR0lMHBwWZHmHN2797Nnj17AHjkkUf2Pf+SJEmSDl4VLXUvAn6UmT8GiIjPAG8A7mhmiN7eXtasWcPo6CgdHR309vY2c/Oz0nQta9deey2bNm1idHSU+fPn093d3RKto7W0MAL7lhk7DXMytjJKkiSpmaoY0mAx8NNx05vLefuJiPMjYn1ErN+6dWvdQ/T39xMRAMybN49ly5bVfRvaX7u3ji5YsIAFCxZUHUOSJEnaT8t2lJKZVwNXA/T09GS919/Z2cnSpUu56aab6Ovrs5OUJmjV1lFb1SRJktTOqmip2wI8fdz0SeW8puvv7+eUU06xla5JbB2VJEmS6q+Kou5bwLMj4uSIOBQ4G5j5yNAHobOzkxUrVthK1yRjraMRYeuoJEmSVCdNP/0yM0cj4kLg3yiGNLg2M7/f7ByqRn9/P5s3b7aVTpIkSaqTGBurrZX19PTk+vXrq44hSZIkSZWIiFszs2ei+6o4/VKSJEmSVCcWdZIkSZLUxizqJEmSJKmNWdRJkiRJUhtri45SImIrMNSg1R8HPNigdTdau2Zv19zQvtnbNTe0b/Z2zQ3tm71dc0P7Zm/X3NC+2ds1N7Rv9nbNDe2bvV1zQ2Ozd2fm8RPd0RZFXSNFxPrJepFpde2avV1zQ/tmb9fc0L7Z2zU3tG/2ds0N7Zu9XXND+2Zv19zQvtnbNTe0b/Z2zQ3VZff0S0mSJElqYxZ1kiRJktTGLOrg6qoDzEC7Zm/X3NC+2ds1N7Rv9nbNDe2bvV1zQ/tmb9fc0L7Z2zU3tG/2ds0N7Zu9XXNDRdnn/DV1kiRJktTObKmTJEmSpDZmUSdJkiRJbWxOFXURkRHxj+OmOyJia0R8qcpck5lteSPi9RHxruoS1vc5jYhFEfGH9U144CJiT0R8JyK+HxHfjYg/iYi2eG+Py357RPxTRCyqOtPBaPX/QbvtS6bT6o/nYPNFRE9EfKjxCacWEdtn8LebIuK4euaph4j4s/L9uaF8r54RER+LiOdWnW28Gj5Hz42IK8vb742IP60q62QioisiPhMRd0fErRHxLxHxnEmWXRIRtzc747jt15x1mvW8IyIOa4UsU6z/zIj4r/Va3wTrP6j9xkT7jKpfF+O1+uf7eC0ZqoF2AM+PiIXl9KuALRMtGBEdTUs1uVmVNzNvzMzLK0n2uJqf0xosAg6oqItCvd93OzPz9Mx8HsXjeS3w53XeRqOMZX8+8DBwQdWBDlJN/4MK36f1fN23glZ/PAeVLzPXZ+bFtW6kRfb7LS8iXgK8DvjVzDwNeCXw08z8g8y8Y4Ll5zc74zit/tqeUkQEMACsycxnZuYLgXcDJ1Sb7MnqnPUdwEEXdU163s4EGlbUtYOD3Ge2zXesuVbUAfwL8Ovl7TcDnx67ozzq9YmIGAQ+UUW4CdScNyKeFxH/WR5R2BARz26xvOOPML6xbJ35bkSsK+c1K/9UGV8UEd+MiNsi4n9HxClTZLsceGY57wPlcv9fRHyrXOZ95bwlEbExIlYBtwNPb9DjIjMfAM4HLiwLyKdExMcj4nvlY3p5menciFgdEV+OiLsi4q8blekAfBNYDBARzyyz3RoR34iIUyPi6IgYGiuKI+LwiPhpRBxSaeonmOB/cG5E3BgRXwNuqjDadPuS68rneigilkXEX5evmy+PPccR8cKIWFv+X/4tIk6s4HGMmerxLC3fl98pX/dHtli+yfYzZ8bjLTLHRMQXy33JzRFxWjm/ks+piPiNiLilzPzvEXFCOf/YiPhKFEexPwZEOf/SiHjHuL9/f0S8vVl5n+BE4MHM3AWQmQ9m5r0RsSYiesp82yPibyPiu8BLKso5ZtLXzmQm2mc2NOHkXg7szsy/H5uRmd8F/iMiPlB+7n8vIn6nonzjTZb1toi4KSK+XWZ9A+z7LP9BRHwyIu6MiM9HxGERcTHwNODrEfH1OmeZ8Hkbv68op6+MiHPL25si4n3j8p8aEUuAtwF/VO4XXxYRx0fEF6L4zvKtiOgt//5J+8+IODEi1sXjZ9a8bKIHUX7mTZZ3TfmcjT2H8YS/XRgR/xoR/085a35EfLTct3wlygMdk73Wp3g8ddtn1uM7VkR8JCLWl4/rfTPJM1HAOfMDbAdOAz4PPAX4DsWRiy+V978XuBVYWHXWg8kLrATeUt4+tNmPo4a85wJXlre/Bywuby9qVv4aMh4FdJS3Xwl8YbJswBLg9nHrfjVFN7ZBccDkS0Bfudxe4MWNet4nmLeN4gjfnwDXlvNOBX5SPu5zgR8DR5fTQ8DTq3iNl7/nA58DXlNO3wQ8u7x9BvC18vYNwMvL278DfKzZmQ/if3AusBk4psp8NexL/gM4BHgB8DPgteV9A8Bvlvf9b+D4cc//tS36eP4J6C1vHzH2nm6hfJPtZ8YvsxL48/L2K4DvjPtfNfRzapLXcyeP95j9B8Dflrc/BKwob/86kMBxFPu9b5fz5wF3A8dW9Ho5ovwf/BD4MLC0nL8G6ClvJ/CmKvId4GvnXB7/HH0v8Kfl7Qn3mRXkvxj4uwnm/xbwVYp9/QkUn0Un8oTP0RbJ2gEcVd4+DvgRxef6kvJ1MrZvuXbc878JOK6Jz9u+10S53JXAueOyXFTe/kPKz8nxr5dy+lPAS8vbzwDuLG8/af9J8V3iz8p584EjJ3ntTpX3EeAkiv3BN8dte1P53P47sLyctwQYBU4vpz8L/N5Ur/UpHs97mcE+kzp/x6L8LlA+R2uA0+r1mp5zp25k5obyiMWbKY6GPdGNmbmzqaGmcIB5vwn8WUScBKzOzLuak/JxNeQdMwj8Q0R8FlhdzmtK/mkyHg1cF0VLXFJ8kZ0w2xMOMkFR1L0auK2cPgJ4NsWbfCgzb67zQ6nFSym+HJKZP4iIIWDsHP2bMvMRgIi4A+gGftrkfAsj4jsULXR3Al+NiCMoThH53LjneEH5+3qKYuLrwNkUX9DawVcz8+EqA9Tw3vzXzNwdEd+j+LD5cjn/exQfsKcAz6f4H1Euc19jU09umsczCPzPiPgkxft1c5PjHex+ZryXUnxBIjO/FkWL2FHlfVV8Tp0EXB9F6+yhwD3l/D5gWZnznyNipLy9KSIeiohfofjyc1tmPtTkzJRZtkfEC4GXUbSIXB9Pvr57D/CFpoebwAF8jgIwzT6zVbwU+HRm7gHuj4i1wK8BG6qNNaEA/jIi+igOyC7m8dMgf5qZg+Xtf6Qoxv6mgVkme94enebvxr5X3Ur5/pzAK4HnjnvNHFW+lp60/4yIbwHXRnHWxhcz8zsHkfc/x/bF5ef+EoqDiVAcsP3rzPzkuHXdM247twJLpnmtT/Z4oHH7zIP5jvWmiDifolg+EXgudXofzMXTLwFupHgTTnRKw44mZ6lFTXkz81PA64GdwL9ExCuaE+9JpsoLQGa+DfgfFKci3hoRxzY5/2QZLwO+nsU1Xr9BcYSl1uc2gL/K4tzr0zPzWZl5TXlf015XEfFfKL6gPDDNorvG3d4DlRzk2ZmZp1Ps7ILimrp5wLZxz+PpmflL5fI3Aq+JiGOAFwJfqyDztCb4H7TKfmWq9+bYqWl7KU4DGhvEdC/FayOA74/7n/xyZr66GaGnMOHjyeLa3T+gaFEfjOpORTug/cwBqOL1tJKiheiXgf+X2jJ/jOKI9f9N0apRmczck5lrMvPPgQspC+Zxfl5+EW0V036OjjPVPrPZvk+xb24Hk2V9C3A88MLy8+l+Hn+9P3Fw53oN9nygz9so+3+Hf+L7cezzfarP9nkUZxCNvWYWZ+b2ifafmbmO4gDOFooD8ssPIOsTM02Ua5Disz2mWX6q1/qEj6e8r277zJl8x4qIk4E/Bc7K4vref+bA9/+TmqtF3bXA+zLze1UHqVFNecsX2o8z80MURz1Oa0a4CUybNyKemZm3ZOYKYCvw9Cbnnyzj0Tx+Ufq54/JOlO0xYPy1Ov8GnDd2ZCgiFkfEUxsTf2IRcTzw9xRfvhL4BsUHFFH0ovUMYGMzM9UiM39GccTzTyhO/bsnIt4I+87Rf0G53HbgW8AHKU49aaUvYcCE/4NWMpN930bg+Cg6nSAiDomI59U13YGb8PGU+5fvZeYVFK+Xqoq6A9rPPMH49+6ZFNeETXd0vpHGZz5n3Px1wO8CRMRrKU7THDMAvIbiSP2/NSHjhCLilNj/Gu3TKU6HamU1v1fL18WE+8wKfA1YULZEUOY5jeJ0td+JiPnlPrIP+M+KMo6ZLGs38EB55sLLy+kxzxjbB1K87sdamp74faBeWSZ73oYoWqUWRNFr9Fk1bOOJGb8CXDRue6eXv5+0/4yIbuD+zPwoxcGaX51kG9+YJO90VgAjwFVTLTTNa33Cx1NPdfiOdRRFgflIFNclv7ae+eZkUZeZm8sv523hAPK+Cbi9bNZ+PrCqsckmVmPeD0RxYentFNfpfJcm5p8i418DfxURt7H/UaQnZStPJRqM4oLgD2TmVyjO6f5meQrb55nZTr5WC6PsbpfinPSvAGMX334YmFfmuZ7inPtdk6ynUpl5G8UpCG+m2Em+NYpOC74PvGHcotcDv1f+bhVT/Q9axkz2fZn5C+C3gSvK/8t3qLgntSkezzvK9+UGYDfwr02OBhzUfgYeP/L/XuCF5WO4nP0LqUY7LCI2j/v54zLP5yLiVuDBccu+D+grX/vLKE43B/a9Zr4OfLbiAzBHUJzuekf5fD6X4vG0rIN4r061z2ya8otuP/DKKLrm/z7wVxSfjRsoPuu/Bvz3zByuIuOYKbL+C9BTfm4uB34w7s82AhdExJ0UBzA+Us6/GvhyHGRHKQf6vGXmTymuM7u9/H3bxGvezz8B/eVn1csoDqT2RNEZ0x0UHanAxPvPM4Hvlvus36E4sLpPFL1K7qI4kHOw/+e3U3yWTtdx22Sv9ckez0zV7TtWlh3xULymPkXRQlk3Yxc9S5KkCkXEbwGvz8xmFnANE0VPtd8G3pgVXOMt1VMU1zl+qTxtWuOUrWUfzcwXVZ1lLpuTLXWSJLWSiHg98H7gf1WdpR6iGNT7RxSdBVjQSbNURLyN4trP/1F1lrnOljpJkiRJamO21EmSJElSG7OokyRJkqQ2ZlEnSZIkSW3Mok6SNCdFxLFlV9XfiYjhiNhS3t4eER8ulzkzIv7ruL95b0T8aXWpJUl6sslGmZckaVYrx5ocG3D3vcD2zPybJyx2JrCdYjxNSZJaki11kiSNU7bOfakcl+ptwB+NG7B3/HLPjIgvR8StEfGNiDi1irySJNlSJ0nSBDJzU0T8PeNa8CLirHGLXA28LTPviogzgA8Dr6ggqiRpjrOokyTpAEXEEcB/BT4XEWOzF1SXSJI0l1nUSZJ04OYB2zLz9KqDSJLkNXWSJE3uMeDIJ87MzEeBeyLijQBReEGzw0mSBBZ1kiRN5Z+A/ok6SgHeArw1Ir4LfB94Q9PTSZIERGZWnUGSJEmSdJBsqZMkSZKkNmZRJ0mSJEltzKJOkiRJktqYRZ0kSZIktTGLOkmSJElqYxZ1kiRJktTGLOokSZIkqY39H5FuEY3hvQjQAAAAAElFTkSuQmCC\n",
            "text/plain": [
              "<Figure size 1080x288 with 1 Axes>"
            ]
          },
          "metadata": {
            "tags": [],
            "needs_background": "light"
          }
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "Kv-Mf5UYZWC-"
      },
      "source": [
        "#split the training and testing data before imputing\n",
        "cleaned_train = cleaned[cleaned[\"PassengerId\"].isin(train[\"PassengerId\"])]i\n",
        "cleaned_test =  cleaned[cleaned[\"PassengerId\"].isin(test[\"PassengerId\"])]"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "NZeBm7TzZWGT"
      },
      "source": [
        "#impute missing ages with the median age per class and title\n",
        "cleaned_train[\"Age\"] = cleaned_train[\"Age\"].fillna(cleaned_train.groupby([\"Pclass\", \"Title\"])[\"Age\"].transform(\"median\"))\n",
        "cleaned_test[\"Age\"] = cleaned_test[\"Age\"].fillna(cleaned_train.groupby([\"Pclass\", \"Title\"])[\"Age\"].transform(\"median\"))"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "ZQep-7PtcGMO"
      },
      "source": [
        "**2.1.2 FARE**"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "VDkjLkxJZWJE",
        "outputId": "e474e2d7-11ae-42cf-dcec-5a4797d1ffa5",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 81
        }
      },
      "source": [
        "# We don't know the fare paid by a single third class passenger\n",
        "missing = raw[raw[\"Fare\"].isna()]\n",
        "missing"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/html": [
              "<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>PassengerId</th>\n",
              "      <th>Survived</th>\n",
              "      <th>Pclass</th>\n",
              "      <th>Name</th>\n",
              "      <th>Sex</th>\n",
              "      <th>Age</th>\n",
              "      <th>SibSp</th>\n",
              "      <th>Parch</th>\n",
              "      <th>Ticket</th>\n",
              "      <th>Fare</th>\n",
              "      <th>Cabin</th>\n",
              "      <th>Embarked</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>152</th>\n",
              "      <td>1044</td>\n",
              "      <td>NaN</td>\n",
              "      <td>3</td>\n",
              "      <td>Storey, Mr. Thomas</td>\n",
              "      <td>male</td>\n",
              "      <td>60.5</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>3701</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>S</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>"
            ],
            "text/plain": [
              "     PassengerId  Survived  Pclass  ... Fare Cabin  Embarked\n",
              "152         1044       NaN       3  ...  NaN   NaN         S\n",
              "\n",
              "[1 rows x 12 columns]"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 38
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "0IKONnZ0ZWM5",
        "outputId": "756f01ff-334a-49d3-a01d-b6ef87494bc0",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 896
        }
      },
      "source": [
        "#check the median and mean Fare per class and title (only of training data to avoid leakage)\n",
        "median = cleaned_train.groupby([\"Pclass\", \"Title\"])[\"Fare\"].median()\n",
        "mean = cleaned_train.groupby([\"Pclass\", \"Title\"])[\"Fare\"].mean()\n",
        "fare = pd.concat([median, mean], axis=1, keys = [\"Median Fare\", \"Mean Fare\"])\n",
        "fare"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/html": [
              "<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></th>\n",
              "      <th>Median Fare</th>\n",
              "      <th>Mean Fare</th>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>Pclass</th>\n",
              "      <th>Title</th>\n",
              "      <th></th>\n",
              "      <th></th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th rowspan=\"15\" valign=\"top\">1</th>\n",
              "      <th>Capt</th>\n",
              "      <td>71.00000</td>\n",
              "      <td>71.000000</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>Col</th>\n",
              "      <td>31.02500</td>\n",
              "      <td>31.025000</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>Countess</th>\n",
              "      <td>86.50000</td>\n",
              "      <td>86.500000</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>Don</th>\n",
              "      <td>27.72080</td>\n",
              "      <td>27.720800</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>Dr</th>\n",
              "      <td>39.60000</td>\n",
              "      <td>63.935840</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>Jonkheer</th>\n",
              "      <td>0.00000</td>\n",
              "      <td>0.000000</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>Lady</th>\n",
              "      <td>39.60000</td>\n",
              "      <td>39.600000</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>Major</th>\n",
              "      <td>28.52500</td>\n",
              "      <td>28.525000</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>Master</th>\n",
              "      <td>120.00000</td>\n",
              "      <td>117.802767</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>Miss</th>\n",
              "      <td>91.75000</td>\n",
              "      <td>124.402715</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>Mlle</th>\n",
              "      <td>59.40210</td>\n",
              "      <td>59.402100</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>Mme</th>\n",
              "      <td>69.30000</td>\n",
              "      <td>69.300000</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>Mr</th>\n",
              "      <td>42.40000</td>\n",
              "      <td>68.034385</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>Mrs</th>\n",
              "      <td>79.42500</td>\n",
              "      <td>93.170636</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>Sir</th>\n",
              "      <td>56.92920</td>\n",
              "      <td>56.929200</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th rowspan=\"7\" valign=\"top\">2</th>\n",
              "      <th>Dr</th>\n",
              "      <td>12.25000</td>\n",
              "      <td>12.250000</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>Master</th>\n",
              "      <td>26.00000</td>\n",
              "      <td>27.306022</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>Miss</th>\n",
              "      <td>13.00000</td>\n",
              "      <td>20.393750</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>Mr</th>\n",
              "      <td>13.00000</td>\n",
              "      <td>19.252564</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>Mrs</th>\n",
              "      <td>26.00000</td>\n",
              "      <td>23.496139</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>Ms</th>\n",
              "      <td>13.00000</td>\n",
              "      <td>13.000000</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>Rev</th>\n",
              "      <td>13.00000</td>\n",
              "      <td>18.312500</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th rowspan=\"4\" valign=\"top\">3</th>\n",
              "      <th>Master</th>\n",
              "      <td>28.51250</td>\n",
              "      <td>28.177232</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>Miss</th>\n",
              "      <td>8.75625</td>\n",
              "      <td>15.248043</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>Mr</th>\n",
              "      <td>7.89580</td>\n",
              "      <td>11.299762</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>Mrs</th>\n",
              "      <td>15.97500</td>\n",
              "      <td>18.233529</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>"
            ],
            "text/plain": [
              "                 Median Fare   Mean Fare\n",
              "Pclass Title                            \n",
              "1      Capt         71.00000   71.000000\n",
              "       Col          31.02500   31.025000\n",
              "       Countess     86.50000   86.500000\n",
              "       Don          27.72080   27.720800\n",
              "       Dr           39.60000   63.935840\n",
              "       Jonkheer      0.00000    0.000000\n",
              "       Lady         39.60000   39.600000\n",
              "       Major        28.52500   28.525000\n",
              "       Master      120.00000  117.802767\n",
              "       Miss         91.75000  124.402715\n",
              "       Mlle         59.40210   59.402100\n",
              "       Mme          69.30000   69.300000\n",
              "       Mr           42.40000   68.034385\n",
              "       Mrs          79.42500   93.170636\n",
              "       Sir          56.92920   56.929200\n",
              "2      Dr           12.25000   12.250000\n",
              "       Master       26.00000   27.306022\n",
              "       Miss         13.00000   20.393750\n",
              "       Mr           13.00000   19.252564\n",
              "       Mrs          26.00000   23.496139\n",
              "       Ms           13.00000   13.000000\n",
              "       Rev          13.00000   18.312500\n",
              "3      Master       28.51250   28.177232\n",
              "       Miss          8.75625   15.248043\n",
              "       Mr            7.89580   11.299762\n",
              "       Mrs          15.97500   18.233529"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 40
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "3qxoadLzZWXX"
      },
      "source": [
        "#impute missing fares by median age per class and title\n",
        "cleaned_train[\"Fare\"] = cleaned_train[\"Fare\"].fillna(cleaned_train.groupby([\"Pclass\", \"Title\"])[\"Fare\"].transform(\"median\"))\n",
        "cleaned_test[\"Fare\"] = cleaned_test[\"Fare\"].fillna(cleaned_train.groupby([\"Pclass\", \"Title\"])[\"Age\"].transform(\"median\"))"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "3FWympGGdab0"
      },
      "source": [
        "####**2.2 BINARY FEATURES**"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "_R3xckZzZWi9",
        "outputId": "16f8fa73-3586-4d8f-ed35-f428dffc9483",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 72
        }
      },
      "source": [
        "# We have no missing binary data to impute\n",
        "print(raw[binary].isna().sum().sort_values(ascending=False))"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "Survived    418\n",
            "Sex           0\n",
            "dtype: int64\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "Bbez9WhrdwRy"
      },
      "source": [
        "####**2.3 DISCRETE FEATURES**"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "4zI57DIBd9Z2",
        "outputId": "ddbf746b-cabd-4eaa-d56f-9e226c13063e",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 109
        }
      },
      "source": [
        "# Perfect, no missing data!\n",
        "print(raw[discrete].isna().sum().sort_values(ascending=False))"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "Parch          0\n",
            "SibSp          0\n",
            "Pclass         0\n",
            "PassengerId    0\n",
            "dtype: int64\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "O-GVBsY9eGQP"
      },
      "source": [
        "####**2.4 NOMINAL FEATURES**"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "dm6q_DtheCaU",
        "outputId": "23200af5-9c71-4a05-d587-57e9703911cd",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 91
        }
      },
      "source": [
        "# number of missing nominal data: \n",
        "print(raw[nominal].isna().sum().sort_values(ascending=False))"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "Embarked    2\n",
            "Ticket      0\n",
            "Name        0\n",
            "dtype: int64\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "t7HbGiwDZWq5",
        "outputId": "d74d5fd2-3e9b-4347-8093-69d388b28c06",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 112
        }
      },
      "source": [
        "# We don't know where two adult female first class passengers boarded the Titanic\n",
        "missing = raw[raw[\"Embarked\"].isna()]\n",
        "missing"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/html": [
              "<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>PassengerId</th>\n",
              "      <th>Survived</th>\n",
              "      <th>Pclass</th>\n",
              "      <th>Name</th>\n",
              "      <th>Sex</th>\n",
              "      <th>Age</th>\n",
              "      <th>SibSp</th>\n",
              "      <th>Parch</th>\n",
              "      <th>Ticket</th>\n",
              "      <th>Fare</th>\n",
              "      <th>Cabin</th>\n",
              "      <th>Embarked</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>61</th>\n",
              "      <td>62</td>\n",
              "      <td>1.0</td>\n",
              "      <td>1</td>\n",
              "      <td>Icard, Miss. Amelie</td>\n",
              "      <td>female</td>\n",
              "      <td>38.0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>113572</td>\n",
              "      <td>80.0</td>\n",
              "      <td>B28</td>\n",
              "      <td>NaN</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>829</th>\n",
              "      <td>830</td>\n",
              "      <td>1.0</td>\n",
              "      <td>1</td>\n",
              "      <td>Stone, Mrs. George Nelson (Martha Evelyn)</td>\n",
              "      <td>female</td>\n",
              "      <td>62.0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>113572</td>\n",
              "      <td>80.0</td>\n",
              "      <td>B28</td>\n",
              "      <td>NaN</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>"
            ],
            "text/plain": [
              "     PassengerId  Survived  Pclass  ...  Fare Cabin  Embarked\n",
              "61            62       1.0       1  ...  80.0   B28       NaN\n",
              "829          830       1.0       1  ...  80.0   B28       NaN\n",
              "\n",
              "[2 rows x 12 columns]"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 49
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "-pp4gD4vebVA",
        "outputId": "97ebde09-54e9-412c-e87e-e354e489ec05",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 295
        }
      },
      "source": [
        "#Most first class passengers boarded in Southampton\n",
        "ax = sns.countplot(x=\"Pclass\", hue=\"Embarked\", data=raw, palette = 'Set2')\n",
        "ax.set(title = \"Embarkation port by Class\")\n",
        "plt.show()"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<Figure size 432x288 with 1 Axes>"
            ]
          },
          "metadata": {
            "tags": [],
            "needs_background": "light"
          }
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "4VXZOs6VeurV"
      },
      "source": [
        "#impute missing values by Southampton\n",
        "cleaned_train[\"Embarked\"].fillna(value = \"S\", inplace = True)\n",
        "cleaned_test[\"Embarked\"].fillna(value = \"S\", inplace = True)"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "XElZ-yHWe9KO"
      },
      "source": [
        "####**2.5 ORDINAL FEATURES**"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "tpnGSZa_fEDU"
      },
      "source": [
        "Note: I think Cabin is a discrete feature because it indicates the Deck the passenger was on. My hypothesis is that people on higher decks had a greater chance of survival. We will check later whether my assumption was correct."
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "mdgRF3AZe_yG",
        "outputId": "7e039fe8-829c-4758-96cd-d584a2291f78",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 54
        }
      },
      "source": [
        "# The cabin of a lot of passengers is missing\n",
        "print(raw[ordinal].isna().sum().sort_values(ascending=False))"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "Cabin    1014\n",
            "dtype: int64\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "f9e7k84OfXjD"
      },
      "source": [
        "#impute missing cabin numbers by unknown\n",
        "cleaned_train[\"Cabin\"].fillna(value = \"U\", inplace = True)\n",
        "cleaned_test[\"Cabin\"].fillna(value = \"U\", inplace = True)"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "xp9d5ax7fXr9"
      },
      "source": [
        "# After imputing we can put the training and testing data back together\n",
        "cleaned = pd.concat([cleaned_train, cleaned_test])\n",
        "#cleaned.to_csv(\"cleaned2.csv\")"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "JCJSCttJfzqb",
        "outputId": "7c78fc24-13b0-4823-867e-0bbfba01d014",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 276
        }
      },
      "source": [
        "# Double-check if there is no more missing data\n",
        "print(cleaned.isna().sum().sort_values(ascending=False))"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "Survived       418\n",
            "Title            0\n",
            "Embarked         0\n",
            "Cabin            0\n",
            "Fare             0\n",
            "Ticket           0\n",
            "Parch            0\n",
            "SibSp            0\n",
            "Age              0\n",
            "Sex              0\n",
            "Name             0\n",
            "Pclass           0\n",
            "PassengerId      0\n",
            "dtype: int64\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "rQIOjFB6mJhS"
      },
      "source": [
        "## 3 E.D.A & FEATURE ENGINEERING"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "08lPNJNHgGvV"
      },
      "source": [
        "Now the fun part starts. Let's see what we can find! Before we continue, I will make a copy of the cleaned version of the dataset."
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "Lp37AWOrp6x8",
        "outputId": "0beb20a3-a1b8-432b-ffd8-7334c0f88086",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 144
        }
      },
      "source": [
        "#We can get a first glimpse of the characteristics of survival by plotting a simple pivot table\n",
        "pd.pivot_table(cleaned, index = \"Survived\", values = [\"Age\", \"SibSp\", \"Parch\", \"Fare\"])"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/html": [
              "<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>Age</th>\n",
              "      <th>Fare</th>\n",
              "      <th>Parch</th>\n",
              "      <th>SibSp</th>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>Survived</th>\n",
              "      <th></th>\n",
              "      <th></th>\n",
              "      <th></th>\n",
              "      <th></th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>0.0</th>\n",
              "      <td>29.799636</td>\n",
              "      <td>22.117887</td>\n",
              "      <td>0.329690</td>\n",
              "      <td>0.553734</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1.0</th>\n",
              "      <td>28.098450</td>\n",
              "      <td>48.395408</td>\n",
              "      <td>0.464912</td>\n",
              "      <td>0.473684</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>"
            ],
            "text/plain": [
              "                Age       Fare     Parch     SibSp\n",
              "Survived                                          \n",
              "0.0       29.799636  22.117887  0.329690  0.553734\n",
              "1.0       28.098450  48.395408  0.464912  0.473684"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 122
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "rBpEMtT0mYzp"
      },
      "source": [
        "engineered = cleaned.copy()"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "7do4E2ztg28E"
      },
      "source": [
        "####**3.1 CONTINUOUS FEATURES**"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "EWykjp2gmbTy"
      },
      "source": [
        "**3.1.1 AGE**"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "KhhvW8QH4Axv"
      },
      "source": [
        ""
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "e0-akdRigXRv"
      },
      "source": [
        "#Visualizing the Age feature will be easier if we turn it from a float to an integer\n",
        "engineered[\"Age\"] = cleaned[\"Age\"].astype(\"int\")"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "v-PvW54UgbTk",
        "outputId": "2f057118-b280-49ff-95e7-877d237440d7",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 456
        }
      },
      "source": [
        "# It looks like young people had a much better chance of survival than people in their twenties and thirties.\n",
        "fig, ax = plt.subplots(figsize=(30,7))  \n",
        "sns.countplot(x=\"Age\", data=engineered, hue = \"Survived\", palette = \"Set2\")\n",
        "ax.set(title = 'Number of survivers per Age')\n",
        "plt.legend(loc= \"upper right\")\n",
        "plt.show()"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<Figure size 2160x504 with 1 Axes>"
            ]
          },
          "metadata": {
            "tags": [],
            "needs_background": "light"
          }
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "DWMo9t6Ih4y1"
      },
      "source": [
        "We can see that most passengers were in their early twenties until late thirties. There weren't many children on board, but we know they were favored during evacuation. We can speculate that older passengers had a lower chance of survival.\n",
        "\n",
        "I want to create a feature binning the ages of passenger. I hope this feature will be less noisy than using the actual ages. Tree based models might benefit from this approach."
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "mmPJlr9_mIiP"
      },
      "source": [
        "# Specify the boundaries of the bins\n",
        "bins = [0, 5,  16, 35,   60,  100]\n",
        "\n",
        "#Bin labels\n",
        "labels = [\"Child\", \"Teen\", \" Young Adult\", \"Adult\", \"Senior\"]\n",
        "\n",
        "#bin the Age using the specified boundaries\n",
        "engineered[\"Age_binned\"] = pd.cut(cleaned[\"Age\"], bins, labels = labels)"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "d5jkac0EiQPf",
        "outputId": "b75456ce-9a22-4aad-e05b-3ef629516b82",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 280
        }
      },
      "source": [
        "# Children indeed had a much better chance of surviving\n",
        "sns.barplot(x=\"Age_binned\", y = \"Survived\", data = engineered, palette = 'Set2')\n",
        "ax.set(title = \"Survival Rate by Age Group\")\n",
        "plt.show()"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<Figure size 432x288 with 1 Axes>"
            ]
          },
          "metadata": {
            "tags": [],
            "needs_background": "light"
          }
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "_jGWiREVjEa4"
      },
      "source": [
        "I also want to keep Age as a continuous feature. We will check here if the feature is normally distributed. Scaling will happen further on in the notebook."
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "0vXKQ_1Ki_Jz",
        "outputId": "27a786b6-cc01-46fb-e9db-4ec8603b8819",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 297
        }
      },
      "source": [
        "#Age looks fairly normally distributed\n",
        "sns.distplot(cleaned[\"Age\"], kde=False)"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "<matplotlib.axes._subplots.AxesSubplot at 0x7ff574e5ee48>"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 70
        },
        {
          "output_type": "display_data",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<Figure size 432x288 with 1 Axes>"
            ]
          },
          "metadata": {
            "tags": [],
            "needs_background": "light"
          }
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "zeAH4zIYjaSf",
        "outputId": "54afcd56-0ed4-4dbc-8a61-c1afe2d69539",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 35
        }
      },
      "source": [
        "#check skewness: Age is not normally distributed, but it isn't too far off.\n",
        "cleaned[\"Age\"].skew()"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "0.4818022725950596"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 71
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "-hLaTr62mrGN"
      },
      "source": [
        "**3.1.2 FARE**"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "ULQCJTe1j8_9",
        "outputId": "f5d54178-8072-4053-f73a-8bb7ff54452e",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 295
        }
      },
      "source": [
        "# The fare differs greatly between classes. Within first class there is also great variation.\n",
        "plt.figure(figsize=(6,4))\n",
        "ax =sns.boxplot(x= \"Pclass\", y = \"Fare\", data = cleaned , palette = 'Set2')\n",
        "ax.set(title = 'Fare by Class')\n",
        "plt.show()"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<Figure size 432x288 with 1 Axes>"
            ]
          },
          "metadata": {
            "tags": [],
            "needs_background": "light"
          }
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "OGJMk2otkexN",
        "outputId": "282a4a4e-6211-4de0-a43d-828387ac5a85",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 175
        }
      },
      "source": [
        "# The fare of different classes overlap. Fare looks like a messy feature.\n",
        "median = cleaned_train.groupby([\"Pclass\"])[\"Fare\"].median()\n",
        "mean = cleaned_train.groupby([\"Pclass\"])[\"Fare\"].mean()\n",
        "minimum = cleaned_train.groupby([\"Pclass\"])[\"Fare\"].min()\n",
        "maximum = cleaned_train.groupby([\"Pclass\"])[\"Fare\"].max()\n",
        "FareClass = pd.concat([minimum, mean, median, maximum], axis=1, keys = [\"Min\", \"Mean\", \"Median\", \"Max\"])\n",
        "FareClass"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/html": [
              "<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>Min</th>\n",
              "      <th>Mean</th>\n",
              "      <th>Median</th>\n",
              "      <th>Max</th>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>Pclass</th>\n",
              "      <th></th>\n",
              "      <th></th>\n",
              "      <th></th>\n",
              "      <th></th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>1</th>\n",
              "      <td>0.0</td>\n",
              "      <td>84.154687</td>\n",
              "      <td>60.2875</td>\n",
              "      <td>512.3292</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2</th>\n",
              "      <td>0.0</td>\n",
              "      <td>20.662183</td>\n",
              "      <td>14.2500</td>\n",
              "      <td>73.5000</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>3</th>\n",
              "      <td>0.0</td>\n",
              "      <td>13.675550</td>\n",
              "      <td>8.0500</td>\n",
              "      <td>69.5500</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>"
            ],
            "text/plain": [
              "        Min       Mean   Median       Max\n",
              "Pclass                                   \n",
              "1       0.0  84.154687  60.2875  512.3292\n",
              "2       0.0  20.662183  14.2500   73.5000\n",
              "3       0.0  13.675550   8.0500   69.5500"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 76
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "tgSVWGrfmrrl"
      },
      "source": [
        "# We will bin the fare in order to make it a little less messy\n",
        "bins = [-1, 8.05, 14.25, 32, 60, 250, np.inf]\n",
        "\n",
        "#Bin labels\n",
        "labels = [\"Very Low\", \"Low\", \"Average\",\"Above Average\", \"High\", \"Very High\"]\n",
        "\n",
        "#bin the Age using the specified boundaries\n",
        "engineered[\"Fare_binned\"] = pd.cut(cleaned[\"Fare\"], bins, labels = labels)"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "p5_bdAYAlB6z",
        "outputId": "549ea379-d363-4842-bf56-53848a2cb5a9",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 280
        }
      },
      "source": [
        "# We can clearly see that people who paid more, had a better chance of surviving\n",
        "plt.figure(figsize=(8,4))\n",
        "sns.barplot(x=\"Fare_binned\", y = \"Survived\", data = engineered, palette = 'Set2')\n",
        "ax.set(title = \"Survival Rate by Age Group\")\n",
        "plt.show()"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "image/png": "iVBORw0KGgoAAAANSUhEUgAAAfEAAAEHCAYAAAC++X95AAAABHNCSVQICAgIfAhkiAAAAAlwSFlzAAALEgAACxIB0t1+/AAAADh0RVh0U29mdHdhcmUAbWF0cGxvdGxpYiB2ZXJzaW9uMy4yLjIsIGh0dHA6Ly9tYXRwbG90bGliLm9yZy+WH4yJAAAaC0lEQVR4nO3deZgldX3v8fcH0EtYxIUxI+twFVTigjKiiYgLLqgRNKIwoogSuUYx8UEd8UZJglscr/HRR1Ew190IKNc4USIagoqEbRAEGcSMLJcZ7csiIi5EwO/9o6qHM00vZ5bqM9X9fj1PP+ecqjqnvqf6dH3696s69UtVIUmS+meLURcgSZI2jCEuSVJPGeKSJPWUIS5JUk8Z4pIk9dRWoy5gfe244461aNGiUZchSdKsuOSSS26uqgWTzetdiC9atIgVK1aMugxJkmZFkuunmmd3uiRJPWWIS5LUU4a4JEk9ZYhLktRThrgkST1liEuS1FOdhXiSTya5MckPp5ifJB9OsirJ5Uke31UtkiTNRV22xD8NHDTN/OcCe7Y/xwAf67AWSZLmnM4u9lJV302yaJpFDgE+W82A5hckuX+Sh1TVz7qqSZKkqSxdupSxsTEWLlzIsmXLRl3OUEZ5xbadgRsGHq9up90rxJMcQ9NaZ7fddpuV4iRJ88vY2Bhr1qwZdRnrpRcntlXVKVW1uKoWL1gw6eVjJUmad0YZ4muAXQce79JOkyRJQxhliC8HjmzPUn8ScJvHwyVJGl5nx8STfBF4GrBjktXA3wD3AaiqjwNnAs8DVgG/AV7VVS2SJM1FXZ6dvmSG+QW8vqv1S5I01/XixDZJknRvhrgkST1liEuS1FOGuCRJPWWIS5LUU4a4JEk9ZYhLktRThrgkST1liEuS1FOGuCRJPWWIS5LUU4a4JEk9ZYhLktRThrgkST1liEuS1FOGuCRJPWWIS5LUU4a4JEk9ZYhLktRThrgkST1liEuS1FOGuCRJPWWIS5LUU4a4JEk9ZYhLktRThrgkST1liEuS1FOGuCRJPWWIS5LUU4a4JEk9ZYhLktRThrgkST3VaYgnOSjJ1UlWJTl+kvm7JTknyaVJLk/yvC7rkSRpLuksxJNsCXwUeC6wN7Akyd4TFns7cHpVPQ44HDipq3okSZprumyJ7wesqqprqup3wKnAIROWKeB+7f0dgJ92WI8kSXNKlyG+M3DDwOPV7bRBfwu8PMlq4EzgDZO9UJJjkqxIsuKmm27qolZJknpn1Ce2LQE+XVW7AM8DPpfkXjVV1SlVtbiqFi9YsGDWi5QkaXPUZYivAXYdeLxLO23Q0cDpAFV1PrA1sGOHNUmSNGd0GeIXA3sm2SPJfWlOXFs+YZn/CxwIkOSRNCFuf7kkSUPoLMSr6i7gWOAs4Cqas9CvTHJikoPbxd4EvCbJD4AvAkdVVXVVkyRJc8lWXb54VZ1Jc8La4LQTBu6vBJ7cZQ2SpH766UWze+mQu+/4XXv701lf9077nTnzQpMY9YltkiRpAxnikiT1lCEuSVJPdXpMXJI0ekuXLmVsbIyFCxeybNmyUZejTcgQl6Q5bmxsjDVrJl6mQ3OB3emSJPWUIS5JUk8Z4pIk9ZQhLklSTxnikiT1lCEuSVJPGeKSJPWUIS5JUk8Z4pIk9ZQhLklSTxnikiT1lCEuSVJPGeKSJPWUIS5JUk8Z4pIk9ZQhLklSTxnikiT11FajLkCS5pPPXXjMrK/z9jtuXHs72+t/xRNPmdX1zTe2xCVJ6ilDXJKknjLEJUnqKUNckqSeMsQlSeopQ1ySpJ4yxCVJ6ilDXJKknjLEJUnqqU5DPMlBSa5OsirJ8VMs89IkK5NcmeSfuqxHkqS5ZNrLria5Haip5lfV/aZ57pbAR4FnAauBi5Msr6qVA8vsCbwNeHJV3ZrkwetZv6R5aunSpYyNjbFw4UKWLVs26nKkkZg2xKtqe4Ak7wR+BnwOCHAE8JAZXns/YFVVXdO+xqnAIcDKgWVeA3y0qm5t13fjBrwHSfPQ2NgYa9asGXUZ0kgN251+cFWdVFW3V9Uvq+pjNIE8nZ2BGwYer26nDdoL2CvJeUkuSHLQkPVIkjTvDRviv05yRJItk2yR5Ajg15tg/VsBewJPA5YAn0hy/4kLJTkmyYokK2666aZNsFpJkvpv2BB/GfBS4P+1Py9pp01nDbDrwONd2mmDVgPLq+rOqroW+DFNqK+jqk6pqsVVtXjBggVDlixJ0tw21HjiVXUdM3efT3QxsGeSPWjC+3DuHfz/TNMC/1SSHWm6169Zz/VIkjQvDdUST7JXkrOT/LB9/Jgkb5/uOVV1F3AscBZwFXB6VV2Z5MQkB7eLnQXckmQlcA7wlqq6ZUPfjCRJ88lQLXHgE8BbgJMBqury9jvd75ruSVV1JnDmhGknDNwv4Lj2R5IkrYdhj4lvU1UXTZh216YuRpIkDW/YlvjNSR5Ke+GXJIfSfG9ckrSZ2+4BW61zq7lj2N/o64FTgEckWQNcS3PBF0nSZu7Zr95p1CWoI8OG+PVV9cwk2wJbVNXtXRYlSZJmNuwx8WuTnAI8CfhVh/VIkqQhDRvijwD+jaZb/dokH0myf3dlSZKkmQwV4lX1m6o6var+DHgccD/gO51WJkmSpjX0eOJJnprkJOASYGuay7BKkqQRGerEtiTXAZcCp9NcVW1TDH4iSZI2wrBnpz+mqn7ZaSWSJGm9TBviSZZW1TLg3Ulq4vyq+svOKpMkSdOaqSV+VXu7outCJEnS+pk2xKvqX9q7V1TV92ehHkmSNKRhz07/QJKrkrwzyaM6rUiSJA1l2O+JPx14OnATcHKSK2YaT1ySJHVr6O+JV9VYVX0YeC1wGXDCDE+RtAGWLl3KkUceydKlS0ddiqTN3LDfE38kcBjwYuAW4DTgTR3WJc1bY2NjrFmzZtRlSOqBYb8n/kngVOA5VfXTDuuR1FOrPja7V2K+87bfrr2d7XU/7C+eOqvrk6YyY4gn2RK4tqo+NAv1SJKkIc14TLyq7gZ2TXLfWahHkiQNadju9GuB85IsB9ZeN72q/qGTqiRJ0oyGDfGftD9bANt3V44kSRrWUCFeVX/XdSGSJGn9DPsVs3OAyQZAecYmr0iSJA1l2O70Nw/c35rm++J3bfpyJEnSsIbtTr9kwqTzklzUQT2SJGlIw3anP3Dg4RbAYmCHTiqSJElDGbY7/RLuOSZ+F3AdcHQXBUmSpOFMG+JJngDcUFV7tI9fSXM8/DpgZefVSZI0S3Z8AEDa236YqSV+MvBMgCQHAO8F3gDsA5wCHNppddKInfT57836Om+7/Y61t7O9/te9fP9ZXZ+0OXnr0f27MOlMIb5lVf28vX8YcEpVnQGckeSybkvTXLN06VLGxsZYuHAhy5YtG3U5ktR7M4Z4kq2q6i7gQOCY9XiutA6H2JSkTWumIP4i8J0kNwO/Bc4FSPIw4LaOa5MkSdOYdhSzqno38Cbg08D+VTV+hvoWNMfGp5XkoCRXJ1mV5PhplntxkkqyePjSJUma32bsEq+qCyaZ9uOZnteOQ/5R4FnAauDiJMurauWE5bYH/gq4cNiiJUnSEOOJb4T9gFVVdU1V/Q44FThkkuXeCbwPuKPDWiRJmnO6DPGdgRsGHq9up62V5PHArlX19eleKMkxSVYkWXHTTTdt+kolSeqhLkN8Wkm2AP6B5pj7tKrqlKpaXFWLFyxY0H1xkiT1QJdfE1sD7DrweJd22rjtgUcB304CsBBYnuTgqlrRYV2S5oAHbXP/dW6l+ajLEL8Y2DPJHjThfTjwsvGZVXUbsOP44yTfBt5sgEsaxhsPePmoS5BGrrPu9PYCMccCZwFXAadX1ZVJTkxycFfrlSRpvuj0qmtVdSZw5oRpJ0yx7NO6rEWSpLlmZCe2SZKkjeP1z6XNzDbb3X+dW0maiiEubWae8pwjR12CpJ6wO12SpJ4yxCVJ6im70+epGz+2dNbXefdtN6+9nc31P/gvls3auiRpNtkSlySppwxxSZJ6yhCXJKmnDHFJknrKEJckqacMcUmSesoQlySppwxxSZJ6yhCXJKmnDHFJknrKEJckqacMcUmSesoQlySppwxxSZJ6yhCXJKmnDHFJknpqq1EXoPljwTb3XedWkrRxDHHNmv95wF6jLkGS5hRDfCMtXbqUsbExFi5cyLJly0ZdjiRpHjHEN9LY2Bhr1qwZdRmSpHnIE9skSeopQ1ySpJ4yxCVJ6ilDXJKknjLEJUnqKUNckqSe6jTEkxyU5Ookq5IcP8n845KsTHJ5krOT7N5lPZIkzSWdhXiSLYGPAs8F9gaWJNl7wmKXAour6jHAlwGvliJJ0pC6bInvB6yqqmuq6nfAqcAhgwtU1TlV9Zv24QXALh3WI0nSnNLlFdt2Bm4YeLwaeOI0yx8N/OtkM5IcAxwDsNtuu0270jf962fXq8iNdfNvbl97O9vr/sBzj5zV9UmSNi+bxYltSV4OLAbeP9n8qjqlqhZX1eIFCxbMbnGSJG2mumyJrwF2HXi8SzttHUmeCfw18NSq+q8O65EkaU7psiV+MbBnkj2S3Bc4HFg+uECSxwEnAwdX1Y0d1iJJ0pzTWYhX1V3AscBZwFXA6VV1ZZITkxzcLvZ+YDvgS0kuS7J8ipeTJEkTdDoUaVWdCZw5YdoJA/ef2eX6JUmayzaLE9skSdL6M8QlSeopQ1ySpJ4yxCVJ6qlOT2ybD+57v23XuZUkabYY4hvpoS959qhLkCTNU3anS5LUU4a4JEk9ZYhLktRThrgkST1liEuS1FOGuCRJPWWIS5LUU4a4JEk9ZYhLktRThrgkST1liEuS1FOGuCRJPWWIS5LUU4a4JEk9ZYhLktRThrgkST1liEuS1FOGuCRJPWWIS5LUU4a4JEk9ZYhLktRThrgkST1liEuS1FOGuCRJPWWIS5LUU4a4JEk9ZYhLktRTnYZ4koOSXJ1kVZLjJ5n/35Kc1s6/MMmiLuuRJGku6SzEk2wJfBR4LrA3sCTJ3hMWOxq4taoeBnwQeF9X9UiSNNd02RLfD1hVVddU1e+AU4FDJixzCPCZ9v6XgQOTpMOaJEmaM1JV3bxwcihwUFX9efv4FcATq+rYgWV+2C6zun38k3aZmye81jHAMe3DhwNXd1L0htsRuHnGpQRuq2G5nYbjdhqe22o4m+N22r2qFkw2Y6vZrmRDVNUpwCmjrmMqSVZU1eJR19EHbqvhuJ2G43YanttqOH3bTl12p68Bdh14vEs7bdJlkmwF7ADc0mFNkiTNGV2G+MXAnkn2SHJf4HBg+YRllgOvbO8fCvx7ddW/L0nSHNNZd3pV3ZXkWOAsYEvgk1V1ZZITgRVVtRz438DnkqwCfk4T9H202Xb1b4bcVsNxOw3H7TQ8t9VwerWdOjuxTZIkdcsrtkmS1FOGuCRJPTWvQjzJOUmeM2HaG5N8bBOv59Pt9+TnlSS/GnUNo5bkhUkqySNGXUtXJnuPSZ6W5GuzXMeOSe5M8trZXO/mZOLfXJKjknykvf/aJEfO8Py1y/fFKPfj49s7yU5JvjzEa3S+T5xXIQ58kXufPHd4O31G7aVkpeksAb7X3m6Uzfjztsne40Z6CXDBpqqj/ZrrnFFVH6+qz466jg6MfD9eVT+tqs2ioTbfQvzLwPPbr7zRDriyE3BukmcnOT/J95N8Kcl27TLXJXlfku8Dx7e3tPP2HHw8nSRbJ/lUkiuSXJrk6e30ryd5THv/0iQntPdPTPKaTffWRyPJPkkuSHJ5kq8keUCSBye5pJ3/2LZVt1v7+CdJthlt1Rum/czsTzMmwOHtAEBfGpi/trU65OftJUlek+TiJD9Icsb4tkny0Ha7XpHkXYP/8Sd5S/ucy5P8XZfvccLs+7Wf56uTfDzJFu1zlrR1/jDJ+9ppr03y/oHXHWxBvjzJRUkuS3LyNDvdJcCbgJ2T7JJkhyTXD6x32yQ3JLlPu72+keSSJOeO9yK0ra2PJ7kQWJZkv/b3cmmS/0jy8Ha5bZKcnmRl+zm+MMnidt6kv8tRS/K3Sd7c3n9C+3m4LMn701wtc9xO7bb5zyTLRlTu+hjZfnzgOYvGt+F0n412/rvbv98Lkvzhxr75ieZViFfVz4GLaAZlgWYndDrwIODtwDOr6vHACuC4gafeUlWPr6p3A7cl2aed/irgU0Ou/vVNCfVomp3PZ5JsDZwLPCXJDsBdwJPb5Z8CfHcD3ubm5rPAW6vqMcAVwN9U1Y3A1knuR/M+V9Bsg92BG6vqN6Mrd6McAnyjqn5Mc9GiW4EnJtm2nX8YcGqSHRnu83Yq8H+q6glV9VjgKprwBPgQ8KH287R6/IlJng3sSTN2wT7AvkkO6Oo9Jtl3YN5+wBtoBjx6KPBnSXaiGdjoGW09T0jyQuAM4EUDzx3fNo9s7z+5qvYB7gaOmFhEkl2Bh1TVRTR/w4dV1W3AZcBT28X+FDirqu6k+drQG6pqX+DNwEkDL7cL8CdVdRzwI+ApVfU44ATgPe0yr6MZrGlv4B3Avm0dM/0uu/YHbTBfluQy4MQplvsU8D8GtumgfWi2+aOBw9ptu9ma5f34+yds38lM+tlobQtc0P79fhfY5A2zeRXircGumPEumCfR7HjOa39RrwR2H3jOaQP3/xF4Vds6OAz4pyHXuz/weYCq+hFwPbAXTYgfQBPeXwe2a1tbe1TV5naN+PXS/mNy/6r6TjvpMzTvFeA/aN7zATQ7ygNoAv3c2a5zE1pCM9AP7e1LgG8AL0jTVft84Kus3+ftUW3L8QqaMPujdvofA+Ot/MHP4LPbn0uB7wOPoAn1TWXiexzsyr6oHfDobpq/q/2BJwDfrqqbquou4AvAAVV1E3BNkicleVBb53nAgTQ7wYvbbXMg8N8nqeMwmh33xDpOa+dB8/d9Wtsa+xPgS+1rngw8ZOC1vtTWDM1VI7/UtrI+yD3be//x911VPwQub6fP9Lvs2m+rap/xH5p/PNaR5P7A9lV1fjtp4j7r7Kq6raruAFYyu/VvqNnaj79lwvadzFSfDYDfAePnilwCLJr5ra2fOXUMaEhfBT6Y5PHANlV1SZIXAN+qqqmOrf164P4ZwN8A/w5cUlUbe5nYi4HFwDXAt2guvv8aml/4XPZdmtDeneZ38lagaP6R6Z0kD6RpbT46SdFc4Kho/st/Pc3FjFZU1e1JwvCft08DL6yqHyQ5CnjaTKUA762qkzf0vUz5wlO8xyRvaReZeNGJmS5CcSrwUprW71eqqtpt85mqetsMz10CLEwy3krfKcmeNFeBfE9b6740f6fbAr+YZic8uL3fCZxTVS9qu2m/PUMdM/0u++C/Bu7fTT9yYXPbj0/lzoGrkHaybeddS7yqfgWcA3ySe06EuAB4cpKHwdpjaXtN8fw7aK5C9zGG70qHpoV5RPv6ewG7AVe3w7TeQNNqO79d7s3Mga70tnvz1iRPaSe9AhhvlZ8LvBz4z6r6PU3IPY/mhKk+OhT4XFXtXlWLqmpX4FqaQySPp/nHbLwFO/TnDdge+FmS+7But/IFwIvb+4PHps8CXj1wLHDnJA/e+LcHTP0ex3+/+6W5zPIWNK2b79F0ez41zZnkW9KE7/hn4Cs03fODrfuzgUPHa07ywPYwy1rtttquqnZu61gEvBdY0v59X0xzuOFrVXV3Vf0SuDbJS9rnJ8ljp3iPO3DPGA9HDUw/j+YfDpLsTdP1DOv3uxyJqvoFcHuSJ7aT+nplzLVGuB+fzFSfjVkx70K89UXgse0tbdfeUcAXk1xOE6bTfUXoC8DvgW9Os8zJSVa3P+fTHIPbou0WPQ04qqrG/wM+l+ZY8G/b+7vQz27lbQbe8+okx9F0ab2/3a770B6zq6rraFox4/+sfI+mtXTrCOreFJbQhNKgM2h2mF+jOX73NVjvz9s7gAtpdhQ/Gpj+RuC49vkPA25rX/ubNF2D57eftS/T/COwKUz1HsdbPhcDH6E5dn8tTev6Z8DxNDvcH9C0er7a1npru+zu7bFtqmolzXHNb7bv7Vus2/U9TB2n0fyDONh9egRwdJIfAFfS/PMwmWXAe5NcyrqtppOABUlWAu9qX+O2Ddh3jMrRwCfabuZtaT8vPTcb+/FhTPrZ2MjXHJqXXd0Aac743KGq3jHqWjQ/tedN/Lbtgj6cphU6VTBpI7W9CPepqjuSPBT4N+DhbU/aZi/Jdm3rlSTH05wU+FcjLmukNtV+fNSfjT4c+9isJPkKzZm3zxh1LZrX9gU+0h5D/gXw6hHXM9dtA5zTHtYI8Lq+BHjr+UneRrPPv551DxXMO5t4Pz7Sz4YtcUmSemq+HhOXJKn3DHFJknrKEJckqacMcUmSesoQl3ooyd2D13Rury62qdexdpCHSeb9Y3thi85kYAAPSZPzK2ZSP/12msuITqr9OlraK+RtlKr68419DUkbz5a4NAck2S7J2WmGYLwiySHt9EVphgb9LPBDYNes31ClWyX5QpKrknw59wyF+u3cMxTnrzLJcItphvn8cJohPa9Jsnb85alqSPLXSX6c5HvAwzftVpLmHkNc6qfBISi/AtwBvKgdgvHpwAfaljc0o5idVFV/RBOM6zNU6cPb5z4S+CXNsIsTTTfc4kNoRnn6U+DvYerhUtMMa3p4O+15NCOgSZqG3elSP63Tnd5eLeo9bSD/HtgZ+MN29vVVdUF7f3CoUoDtaAJ1qgF3bqiq89r7nwf+EvhfE5aZONziswbm/XPbfb9yvIU+TQ3b01xv/Tfte1o+9duXBIa4NFccASwA9q2qO5NcB2zdzhscgnF9hyodZnjR6YZbHBzmMgO396ohyRuHrElSy+50aW7YgWYkvDuTPJ1mnPbJrO9Qpbsl+eP2/svYNEPFTlXDd4EXJvmDJNsDL9gE65LmNFvi0tzwBeBf2uFHV7DusKVrVdU3kzySZqhSgF/RDNt54xSvezXw+iSfBFbSjL+8Uaaqoaq+n+Q0miFLb6QZ2lTSNBwARZKknrI7XZKknrI7XZrnkjwIOHuSWQdW1S2zXY+k4dmdLklST9mdLklSTxnikiT1lCEuSVJPGeKSJPXU/wd9cMHPJCwAVgAAAABJRU5ErkJggg==\n",
            "text/plain": [
              "<Figure size 576x288 with 1 Axes>"
            ]
          },
          "metadata": {
            "tags": [],
            "needs_background": "light"
          }
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "GHYDmkZflWBC"
      },
      "source": [
        "I also want to keep Fare as a continuous feature. We will check here if the feature is normally distributed. Scaling will happen further on in the notebook."
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "hWbJdwo8lCEw",
        "outputId": "da9cbd17-91b8-4c76-8a14-afa5dde5435b",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 297
        }
      },
      "source": [
        "#Fare is highly skewed\n",
        "sns.distplot(cleaned[\"Fare\"], kde=False)\n",
        "cleaned[\"Fare\"].skew()"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "4.369596551129241"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 82
        },
        {
          "output_type": "display_data",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<Figure size 432x288 with 1 Axes>"
            ]
          },
          "metadata": {
            "tags": [],
            "needs_background": "light"
          }
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "pvh3A3UZEijp"
      },
      "source": [
        "#Log normalizing Fare makes the feature a lot less skewed\n",
        "engineered[\"FareLog\"] =np.log(cleaned[\"Fare\"]+1)"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "xtd2JPUMjulP",
        "outputId": "4c7ccdd4-18c7-40e8-ef80-c248d05a065e",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 297
        }
      },
      "source": [
        "#FareLog is a lot less skwed\n",
        "sns.distplot(engineered[\"FareLog\"], kde=False)\n",
        "engineered[\"FareLog\"].skew()"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "0.5413054618808321"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 229
        },
        {
          "output_type": "display_data",
          "data": {
            "image/png": "iVBORw0KGgoAAAANSUhEUgAAAXcAAAEGCAYAAACevtWaAAAABHNCSVQICAgIfAhkiAAAAAlwSFlzAAALEgAACxIB0t1+/AAAADh0RVh0U29mdHdhcmUAbWF0cGxvdGxpYiB2ZXJzaW9uMy4yLjIsIGh0dHA6Ly9tYXRwbG90bGliLm9yZy+WH4yJAAATi0lEQVR4nO3df7DldX3f8ecrgKBiXJSbHbq7ZpmEJFWnWcgNmCHNKIwW0cniTOKgRqllunYGWxwzieA/xpnSmmmV1JmGdHXRpQWRoAxbh0YpEK0zAbzgyk9ttgqyO8heFRBiJWF594/72XpY7+49955779nz2edj5s79fj/fz/ec93dgX/vdz/mezydVhSSpLz837gIkScvPcJekDhnuktQhw12SOmS4S1KHjh53AQAnnnhibdy4cdxlSNJEueuuu75fVVPzHTsswn3jxo3MzMyMuwxJmihJHj7YMYdlJKlDhrskdchwl6QOGe6S1CHDXZI6ZLhLUocMd0nqkOEuSR0y3CWpQ4fFN1TVn2vu+O7Qfd9+xitWsBLpyOSduyR1yHCXpA4Z7pLUIcNdkjpkuEtShwx3SeqQ4S5JHRo63JMcleTrSb7Q9k9OckeSXUk+m+QFrf3Ytr+rHd+4MqVLkg5mMXfuFwMPDuz/KXB5Vf0y8DhwYWu/EHi8tV/e+kmSVtFQ4Z5kPfAm4JNtP8BZwPWty3bgvLa9ue3Tjp/d+kuSVsmwd+5/Bvwx8FzbfznwRFU92/Z3A+va9jrgEYB2/MnW/3mSbEkyk2RmdnZ2ieVLkuazYLgneTOwt6ruWs43rqqtVTVdVdNTU1PL+dKSdMQbZuKwM4HfTXIucBzw88B/AtYkObrdna8H9rT+e4ANwO4kRwMvBX6w7JVLkg5qwTv3qrq0qtZX1UbgfODWqnoHcBvwe63bBcCNbXtH26cdv7WqalmrliQd0ijPuX8AeH+SXcyNqW9r7duAl7f29wOXjFaiJGmxFjWfe1X9NfDXbfvbwOnz9PkJ8PvLUJskaYn8hqokdchwl6QOGe6S1CHDXZI6ZLhLUocMd0nqkOEuSR0y3CWpQ4a7JHXIcJekDhnuktQhw12SOmS4S1KHDHdJ6pDhLkkdMtwlqUPDLJB9XJI7k3wjyf1JPtzaP53kO0l2tp9NrT1JPp5kV5J7kpy20hchSXq+YVZiegY4q6qeTnIM8NUk/6Md+6Oquv6A/m8ETmk/ZwBXtN+SpFUyzALZVVVPt91j2s+hFrzeDFzVzrsdWJPkpNFLlSQNa6gx9yRHJdkJ7AVurqo72qHL2tDL5UmObW3rgEcGTt/d2g58zS1JZpLMzM7OjnAJkqQDDRXuVbWvqjYB64HTk7wauBT4NeA3gZcBH1jMG1fV1qqarqrpqampRZYtSTqURT0tU1VPALcB51TVo23o5RngU8DprdseYMPAaetbmyRplQzztMxUkjVt+4XA64Fv7h9HTxLgPOC+dsoO4F3tqZnXAE9W1aMrUr0kaV7DPC1zErA9yVHM/WVwXVV9IcmtSaaAADuBf9X63wScC+wCfgy8e/nLliQdyoLhXlX3AKfO037WQfoXcNHopUmSlspvqEpShwx3SeqQ4S5JHTLcJalDhrskdchwl6QOGe6S1CHDXZI6ZLhLUocMd0nqkOEuSR0y3CWpQ4a7JHXIcJekDhnuktShYVZiOi7JnUm+keT+JB9u7ScnuSPJriSfTfKC1n5s29/Vjm9c2UuQJB1omDv3Z4CzqurXgU3AOW35vD8FLq+qXwYeBy5s/S8EHm/tl7d+kqRVtGC4t0Wwn267x7SfAs4Crm/t25lbRxVgc9unHT+7rbMqSVolQ425JzkqyU5gL3Az8H+AJ6rq2dZlN7Cuba8DHgFox58EXj7Pa25JMpNkZnZ2drSrkCQ9z1DhXlX7qmoTsB44Hfi1Ud+4qrZW1XRVTU9NTY36cpKkAYt6WqaqngBuA34LWJNk/wLb64E9bXsPsAGgHX8p8INlqVaSNJRhnpaZSrKmbb8QeD3wIHMh/3ut2wXAjW17R9unHb+1qmo5i5YkHdrRC3fhJGB7kqOY+8vguqr6QpIHgGuT/Fvg68C21n8b8F+T7AJ+CJy/AnVLkg5hwXCvqnuAU+dp/zZz4+8Htv8E+P1lqU6StCR+Q1WSOmS4S1KHDHdJ6pDhLkkdMtwlqUOGuyR1yHCXpA4Z7pLUIcNdkjpkuEtShwx3SeqQ4S5JHTLcJalDhrskdchwl6QOGe6S1KFhltnbkOS2JA8kuT/Jxa39T5LsSbKz/Zw7cM6lSXYl+VaSf7aSFyBJ+lnDLLP3LPCHVXV3kpcAdyW5uR27vKr+42DnJK9kbmm9VwH/CPifSX6lqvYtZ+GSpINb8M69qh6tqrvb9lPMLY697hCnbAaurapnquo7wC7mWY5PkrRyFjXmnmQjc+up3tGa3pvkniRXJjmhta0DHhk4bTfz/GWQZEuSmSQzs7Oziy5cknRwQ4d7kuOBzwHvq6ofAVcAvwRsAh4FPrqYN66qrVU1XVXTU1NTizlVkrSAocI9yTHMBfvVVfV5gKp6rKr2VdVzwCf46dDLHmDDwOnrW5skaZUM87RMgG3Ag1X1sYH2kwa6vQW4r23vAM5PcmySk4FTgDuXr2RJ0kKGeVrmTOCdwL1Jdra2DwJvS7IJKOAh4D0AVXV/kuuAB5h70uYin5SRpNW1YLhX1VeBzHPopkOccxlw2Qh1SZJG4DdUJalDhrskdchwl6QOGe6S1CHDXZI6ZLhLUocMd0nqkOEuSR0y3CWpQ4a7JHXIcJekDhnuktQhw12SOmS4S1KHDHdJ6tAwKzFtSHJbkgeS3J/k4tb+siQ3J/nb9vuE1p4kH0+yqy2efdpKX4Qk6fmGuXN/FvjDqnol8BrgoiSvBC4BbqmqU4Bb2j7AG5lbWu8UYAtzC2lLklbRguFeVY9W1d1t+yngQWAdsBnY3rptB85r25uBq2rO7cCaA9ZblSStsEWNuSfZCJwK3AGsrapH26HvAWvb9jrgkYHTdre2A19rS5KZJDOzs7OLLFuSdChDh3uS44HPAe+rqh8NHquqYm6h7KFV1daqmq6q6ampqcWcKklawFDhnuQY5oL96qr6fGt+bP9wS/u9t7XvATYMnL6+tUmSVskwT8sE2AY8WFUfGzi0A7igbV8A3DjQ/q721MxrgCcHhm8kSavg6CH6nAm8E7g3yc7W9kHgI8B1SS4EHgbe2o7dBJwL7AJ+DLx7WSuWJC1owXCvqq8COcjhs+fpX8BFI9YlSRqB31CVpA4Z7pLUIcNdkjpkuEtShwx3SeqQ4S5JHTLcJalDhrskdchwl6QOGe6S1CHDXZI6ZLhLUocMd0nqkOEuSR0y3CWpQwvO557kSuDNwN6qenVr+xPgXwL7V7b+YFXd1I5dClwI7AP+TVV9cQXqlsbqmju+u6j+bz/jFStUiTS/Ye7cPw2cM0/75VW1qf3sD/ZXAucDr2rn/HmSo5arWEnScBYM96r6CvDDIV9vM3BtVT1TVd9hbqm900eoT5K0BKOMub83yT1JrkxyQmtbBzwy0Gd3a5MkraKlhvsVwC8Bm4BHgY8u9gWSbEkyk2RmdnZ24RMkSUNbUrhX1WNVta+qngM+wU+HXvYAGwa6rm9t873G1qqarqrpqamppZQhSTqIJYV7kpMGdt8C3Ne2dwDnJzk2ycnAKcCdo5UoSVqsYR6F/AzwWuDEJLuBDwGvTbIJKOAh4D0AVXV/kuuAB4BngYuqat/KlC5JOpgFw72q3jZP87ZD9L8MuGyUoiRJo/EbqpLUIcNdkjpkuEtShwx3SerQgh+oSocTJ+yShuOduyR1yHCXpA4Z7pLUIcNdkjpkuEtShwx3SeqQ4S5JHTLcJalDhrskdchwl6QOGe6S1KEFwz3JlUn2JrlvoO1lSW5O8rft9wmtPUk+nmRXknuSnLaSxUuS5jfMnfungXMOaLsEuKWqTgFuafsAb2Ru3dRTgC3AFctTpiRpMRYM96r6CvDDA5o3A9vb9nbgvIH2q2rO7cCaAxbTliStgqWOua+tqkfb9veAtW17HfDIQL/dre1nJNmSZCbJzOzs7BLLkCTNZ+QPVKuqgFrCeVurarqqpqempkYtQ5I0YKnh/tj+4Zb2e29r3wNsGOi3vrVJklbRUsN9B3BB274AuHGg/V3tqZnXAE8ODN9IklbJgsvsJfkM8FrgxCS7gQ8BHwGuS3Ih8DDw1tb9JuBcYBfwY+DdK1CzJGkBC4Z7Vb3tIIfOnqdvAReNWpQkaTR+Q1WSOmS4S1KHDHdJ6tCCY+7SSrvmju+OuwSpO965S1KHvHOXGv8FoZ545y5JHTLcJalDDsuoaw616Ejlnbskdchwl6QOGe6S1CHH3KUJtpjPFN5+xitWsBIdbgx36Qix2A+X/ctgsjksI0kdGunOPclDwFPAPuDZqppO8jLgs8BG4CHgrVX1+GhlSpIWYzmGZV5XVd8f2L8EuKWqPpLkkrb/gWV4H2liOTau1bYSwzKbge1teztw3gq8hyTpEEYN9wK+lOSuJFta29qBRbG/B6yd78QkW5LMJJmZnZ0dsQxJ0qBRh2V+u6r2JPkF4OYk3xw8WFWVpOY7saq2AlsBpqen5+0jSVqake7cq2pP+70XuAE4HXgsyUkA7ffeUYuUJC3OksM9yYuTvGT/NvAG4D5gB3BB63YBcOOoRUqSFmeUYZm1wA1J9r/ONVX1V0m+BlyX5ELgYeCto5cpSVqMJYd7VX0b+PV52n8AnD1KUZKk0fgNVUnqkOEuSR1y4jDpMOPqUVoOhrukeTllwmRzWEaSOmS4S1KHDHdJ6pBj7pJWlWP5q8M7d0nqkHfukrrhvwp+auLD3UV/V4/PX+tg/H/j8OOwjCR1aOLv3CVpKXofwvHOXZI6ZLhLUocMd0nq0IqFe5Jzknwrya4kl6zU+0iSftaKfKCa5CjgPwOvB3YDX0uyo6oeWIn3k9SnSX3E8nD4sHalnpY5HdjVluIjybXAZsBwX2GT+odB0vJaqXBfBzwysL8bOGOwQ5ItwJa2+3SSby3xvU4Evj9s53cs8U1W2KKu4TDlNYzfpNcPh+k1LDI3VjOTfvFgB8b2nHtVbQW2jvo6SWaqanoZShobr+HwMOnXMOn1g9ewnFbqA9U9wIaB/fWtTZK0ClYq3L8GnJLk5CQvAM4HdqzQe0mSDrAiwzJV9WyS9wJfBI4Crqyq+1fivViGoZ3DgNdweJj0a5j0+sFrWDapqnHXIElaZn5DVZI6ZLhLUocmOtwnfYqDJFcm2ZvkvnHXslRJNiS5LckDSe5PcvG4a1qMJMcluTPJN1r9Hx53TUuV5KgkX0/yhXHXshRJHkpyb5KdSWbGXc9SJFmT5Pok30zyYJLfGlstkzrm3qY4+N8MTHEAvG2SpjhI8jvA08BVVfXqcdezFElOAk6qqruTvAS4CzhvUv47JAnw4qp6OskxwFeBi6vq9jGXtmhJ3g9MAz9fVW8edz2LleQhYLqqDrsvMQ0ryXbgf1XVJ9uTgi+qqifGUcsk37n//ykOqurvgf1THEyMqvoK8MNx1zGKqnq0qu5u208BDzL3DeWJUHOebrvHtJ+Ju+NJsh54E/DJcddypEryUuB3gG0AVfX34wp2mOxwn2+Kg4kJlR4l2QicCtwx3koWpw1n7AT2AjdX1UTV3/wZ8MfAc+MuZAQFfCnJXW16kklzMjALfKoNj30yyYvHVcwkh7sOI0mOBz4HvK+qfjTuehajqvZV1Sbmvkl9epKJGiJL8mZgb1XdNe5aRvTbVXUa8EbgojZsOUmOBk4DrqiqU4G/A8b2WeAkh7tTHBwm2lj154Crq+rz465nqdo/oW8Dzhl3LYt0JvC7bcz6WuCsJP9tvCUtXlXtab/3AjcwN/Q6SXYDuwf+5Xc9c2E/FpMc7k5xcBhoH0huAx6sqo+Nu57FSjKVZE3bfiFzH9B/c7xVLU5VXVpV66tqI3N/Dm6tqj8Yc1mLkuTF7QN52lDGG4CJeoqsqr4HPJLkV1vT2YxxmvOxzQo5qlWe4mBFJPkM8FrgxCS7gQ9V1bbxVrVoZwLvBO5t49YAH6yqm8ZY02KcBGxvT1/9HHBdVU3ko4QTbi1ww9y9AkcD11TVX423pCX518DV7Ybz28C7x1XIxD4KKUk6uEkelpEkHYThLkkdMtwlqUOGuyR1yHCXpA5N7KOQ0qEk2QfcO9B0XlU9NOJrPl1Vx49UmLRKDHf16v+2KQWG1r6Qlaqa5PlZJMBhGR0hkhyf5JYkd7c5wze39o1tTYCrmPtG5IYkf5Tka0nuWWh+9ySbktze+t6Q5ITW/putbWeS/zDJc/ZrMhnu6tULW7DuTHID8BPgLW1iqtcBH2136gCnAH9eVa8CfrXtnw5sAn5jgQmsrgI+UFX/hLlhoA+19k8B72n/eti33BcnLcRhGfXqecMybXKzf9eC+jnmpode2w4/PLA4xxvaz9fb/vHMhf1XDnyDNn/3mqr6cmvaDvxlm6vmJVX1N639GmDiFs/QZDPcdaR4BzAF/EZV/UObQfG4duzvBvoF+PdV9V9WuT5pWTksoyPFS5mb8/wfkrwO+MWD9Psi8C/a/PQkWZfkF+brWFVPAo8n+aet6Z3Al9vUwU8lOaO1n79sVyENyTt3HSmuBv57knuBGQ4yrW9VfSnJPwb+pg3JPw38AXOrNL2ozd6538eAC4C/SPIinj8L4IXAJ5I8B3wZeHL5L0k6OGeFlFZAkuP3r82a5BLmFhG/eMxl6Qjinbu0Mt6U5FLm/ow9DPzz8ZajI4137pLUIT9QlaQOGe6S1CHDXZI6ZLhLUocMd0nq0P8Dz4aSLuf3viEAAAAASUVORK5CYII=\n",
            "text/plain": [
              "<Figure size 432x288 with 1 Axes>"
            ]
          },
          "metadata": {
            "tags": [],
            "needs_background": "light"
          }
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "8ihVlEaLmZq8"
      },
      "source": [
        "####**3.2 BINARY FEATURES**"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "2Y9IuTOumTkD"
      },
      "source": [
        "**3.2.1 SEX**"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "HPjgLRQEmoqS",
        "outputId": "687c3141-b8e3-4d44-fab0-882bc038ccf1",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 295
        }
      },
      "source": [
        "# Woman had a much better chance of surviving than man\n",
        "plt.figure(figsize=(8,4))\n",
        "ax = sns.barplot(x=\"Sex\", y = \"Survived\", data = cleaned, palette = 'Set2')\n",
        "ax.set(title = \"Survival Rate by Sex\")\n",
        "plt.show()"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "image/png": "iVBORw0KGgoAAAANSUhEUgAAAfEAAAEWCAYAAAB2c65HAAAABHNCSVQICAgIfAhkiAAAAAlwSFlzAAALEgAACxIB0t1+/AAAADh0RVh0U29mdHdhcmUAbWF0cGxvdGxpYiB2ZXJzaW9uMy4yLjIsIGh0dHA6Ly9tYXRwbG90bGliLm9yZy+WH4yJAAAaiklEQVR4nO3de5hddX3v8feHIOURuehhLBwSDMUoIlXRgLalihUVbIWeKm3AG+dY82gFPV6geCnHgrfi8VaL1WgpiGK42McT23gi9VpRNIMimCAaAU0iqYMXLuoBg9/zx16Dm3Ey2QNZ2bN23q/nmWf2b63f/q3vHtj57PVba6+VqkKSJHXPTsMuQJIk3TuGuCRJHWWIS5LUUYa4JEkdZYhLktRRhrgkSR1liEtzXJL3JfmbbTDOeUneuC1q2taS3JjkqGHXIXWNIS7dC0mOSPKlJLck+XGSy5Mc1sa2qurFVXVWG2NPSnJSkruS3J7k1iTfSPIns3j+nAnhJK9NckPzWjYkuWjYNUltMcSlWUqyB/CvwHuABwH7AX8L3HEvxkqSufI+/HJVPQDYC3gvsDzJXkOuaVaSvAB4HnBU81oWA58eblVSe+bKPx5SlzwMoKo+WlV3VdUvqupTVXU1QJI3JPnwZOckC5NUkp2b9ueSvCnJ5cDPgVOTjPdvIMkrkqxoHt89DZ7k2v495CQ7J5lI8timfUmSTc0MwReSPHK2L66qfgVcAOwGLGrGPTDJZ5L8KMnNST4yGfBJLgD2Bz7R7P2e1ix/QjNb8dNmz/7IrWz6sCRrk/wkyT8n2bUZ55tJntn3mu/X1HDodGMAq6rqu81r2VRVy/qeu2eSf0pyU5KNSd6YZF6SXZJcleSUpt+8ZnbljNn+/aTtyRCXZu/bwF1Jzk9yTJIH3osxngcsBXYH3gc8PMmivvUnAhdO87yPAif0tZ8O3FxVX2van6QXvA8GvgZ8ZLaFJZkH/Hfgl8D3JhcDbwH+K/AIYAHwBoCqeh7wfeCZVfWAqjo7yX7AvwFvpDdb8WrgY0nGZtj0c5rXcyC9D0qvb5Z/CHhuX79nADdV1denGeMK4PlJTk2yuHkt/c4DNgMPBQ4Fngb8ZVXd2WzjzCSPAE4H5gFvmqFeaegMcWmWqupW4AiggA8AE0lWJPntWQxzXlWtqarNVXUL8H9owrkJ84OAFdM870Lg2CT3b9on0gv2ydrOrarbquoOeiH76CR7DljTE5L8FPh/wP8GnltVP2zGXVdVl1XVHVU1AbwDeNIMYz0XWFlVK6vqV1V1GTBOL4C35B+qan1V/ZheeE5+WPkw8IzmMAb0PgBdMN0AVfVh4BR6HwY+D/wwyV8DNP99ngH8z6r6WfPa3gksaZ77TXofOj5O70PH86rqrhnqlYbOEJfuhaq6tqpOqqr5wCH09lDfNYsh1k9pX8ivQ+tE4ONV9fNptrsOuBZ4ZhPkxzbPnZwCfmuS7ya5FbixedreA9Z0RVXtBTyQ3geIP5xckeS3kyxvpqBvpResM437EOD4Zir9p82HgyOAfWd4Tv/f5Hv0/qZU1Q+Ay4FnNVP4xzDDDENVfaSqjqJ3bP/FwFlJnt7UdD/gpr6a3k9v1mLS+U2/lVX1nRlqleYEQ1y6j6rqW/SmaQ9pFv0MuH9fl32me9qU9mXAWJLH0Avz6abSJ01OqR8HrG2CHXrhfxxwFLAnsLBZnkFex92FVd0OvAR4Xt9x5zc3Nf9uVe1Bb0+7f9ypr2c9cEFV7dX3s1tVvXWGTS/oe7w/8IO+9vnNNo+ndwLexgFexy+r6hLganr/bdbTO/lw776a9qiq/vMG3kvvpMWnJzlia9uQhs0Ql2YpyUFJXpVkftNeQC9Ur2i6XAU8Mcn+zVT2a7Y2ZlX9ErgEeBu9Y8iXzdB9Ob1juS/hnmG/O72Q+hG9DxFvns3rmlLPj4EPApMndu0O3A7c0hzvPnXKU/4T+J2+9ofpzRY8vZkh2DXJkZN/sy14aZL5SR4EvA7o/2rYx4HHAi+nd4x8Wul9Ve6Pk+yeZKckxwCPBL5SVTcBnwLenmSPZv2BSZ7UPPd5wOOAk4CXAecnecAM9UpDZ4hLs3cb8HjgK0l+Ri+8vwm8CqA5/nsRvT3AK+nt2Q3iQnp70ZdU1eYtdWrC6MvA73PPoPsQvWnojcBafv2h4t56F71j0Y+i9xW6xwK30Dth7V+m9H0L8PpmmvrVVbWe3qzAa4EJenvBpzLzvzkX0gvZ64Hv0js+DUBV/QL4GHDANNvud2uzze8DPwXOBl5SVV9s1j8f2IXe3+cnwKXAvkn2b17v86vq9qq6kN4x/HfOsC1p6FI1dRZMkuae5uteD6uq5261s7SD2HnYBUjS1jRT7C+kd2a6pIbT6ZLmtCQvojcd/8mq+sKw65HmEqfTJUnqKPfEJUnqqM4dE997771r4cKFwy5DkqTt4sorr7y5qqa9ZHHnQnzhwoWMj49vvaMkSSMgyfe2tM7pdEmSOsoQlySpo1oN8SRHJ7kuybokp0+zfv8kn03y9SRXJ5npDkeSJKlPayHe3Mf3HHp3HDoYOCHJwVO6vR64uKoOpXc7wPe2VY8kSaOmzT3xw4F1VXV9Vd1J76YNx03pU8DkPYL35J53LZIkSTNoM8T34573B97QLOv3BuC5STYAK4FTphsoydIk40nGJyYm2qhVkqTOGfaJbScA51XVfOAZwAVJfqOmqlpWVYuravHY2LRflZMkaYfTZohvBBb0tec3y/q9ELgYoKq+DOwK7N1iTZIkjYw2L/ayGliU5AB64b0EOHFKn+8DTwHOS/IIeiHufLkkbUOnnXYamzZtYp999uHss88edjnahloL8aranORkYBUwDzi3qtYkORMYr6oVwKuADyR5Bb2T3E4q78giSdvUpk2b2Lhx6kSoRkGrl12tqpX0TljrX3ZG3+O1wB+0WYMkSaNq2Ce2SZKke8kQlySpowxxSZI6yhCXJKmjDHFJkjrKEJckqaMMcUmSOsoQlySpowxxSZI6qtUrtknSIH74j6cNu4SRdtctN9/92791ex78ku1/XXr3xCVJ6ihDXJKkjjLEJUnqKENckqSOMsQlSeooQ1ySpI4yxCVJ6qhWQzzJ0UmuS7IuyenTrH9nkquan28n+Wmb9UiSNEpau9hLknnAOcBTgQ3A6iQrqmrtZJ+qekVf/1OAQ9uqR5KkUdPmnvjhwLqqur6q7gSWA8fN0P8E4KMt1iNJ0khpM8T3A9b3tTc0y35DkocABwCfabEeSZJGylw5sW0JcGlV3TXdyiRLk4wnGZ+YmNjOpUmSNDe1GeIbgQV97fnNsuksYYap9KpaVlWLq2rx2NjYNixRkqTuavMuZquBRUkOoBfeS4ATp3ZKchDwQODLLdYiSTussfvvco/fGh2thXhVbU5yMrAKmAecW1VrkpwJjFfViqbrEmB5VVVbtUjSjuy1T3zYsEtQS1q9n3hVrQRWTll2xpT2G9qsQZKkUTVXTmyTJEmzZIhLktRRhrgkSR1liEuS1FGGuCRJHWWIS5LUUYa4JEkdZYhLktRRhrgkSR1liEuS1FGGuCRJHWWIS5LUUYa4JEkdZYhLktRRhrgkSR1liEuS1FGGuCRJHWWIS5LUUa2GeJKjk1yXZF2S07fQ58+TrE2yJsmFbdYjSdIo2bmtgZPMA84BngpsAFYnWVFVa/v6LAJeA/xBVf0kyYPbqkeSpFHT5p744cC6qrq+qu4ElgPHTenzIuCcqvoJQFX9sMV6JEkaKW2G+H7A+r72hmZZv4cBD0tyeZIrkhw93UBJliYZTzI+MTHRUrmSJHXLsE9s2xlYBBwJnAB8IMleUztV1bKqWlxVi8fGxrZziZIkzU1thvhGYEFfe36zrN8GYEVV/bKqbgC+TS/UJUnSVrQZ4quBRUkOSLILsARYMaXPx+nthZNkb3rT69e3WJMkSSOjtRCvqs3AycAq4Frg4qpak+TMJMc23VYBP0qyFvgscGpV/aitmiRJGiWtfcUMoKpWAiunLDuj73EBr2x+JEnSLAz7xDZJknQvGeKSJHWUIS5JUkcZ4pIkdZQhLklSRxnikiR1lCEuSVJHGeKSJHWUIS5JUkcZ4pIkdZQhLklSRxnikiR1lCEuSVJHGeKSJHWUIS5JUkcZ4pIkdZQhLklSR7Ua4kmOTnJdknVJTp9m/UlJJpJc1fz8ZZv1SJI0SnZua+Ak84BzgKcCG4DVSVZU1dopXS+qqpPbqkOSpFHV5p744cC6qrq+qu4ElgPHtbg9SZJ2KG2G+H7A+r72hmbZVM9KcnWSS5MsmG6gJEuTjCcZn5iYaKNWSZI6Z9gntn0CWFhVjwIuA86frlNVLauqxVW1eGxsbLsWKEnSXNVmiG8E+ves5zfL7lZVP6qqO5rmB4HHtViPJEkjpc0QXw0sSnJAkl2AJcCK/g5J9u1rHgtc22I9kiSNlNbOTq+qzUlOBlYB84Bzq2pNkjOB8apaAbwsybHAZuDHwElt1SNJ0qiZMcST3AbUltZX1R4zPb+qVgIrpyw7o+/xa4DXDFSpJEm6hxlDvKp2B0hyFnATcAEQ4DnAvjM8VZIktWzQY+LHVtV7q+q2qrq1qv4Rv/MtSdJQDRriP0vynCTzkuyU5DnAz9osTJIkzWzQED8R+HPgP5uf45tlkiRpSAY6O72qbsTpc0mS5pSB9sSTPCzJp5N8s2k/Ksnr2y1NkiTNZNDp9A/Q+yrYLwGq6mp6F2+RJElDMmiI37+qvjpl2eZtXYwkSRrcoCF+c5IDaS78kuTZ9L43LkmShmTQy66+FFgGHJRkI3ADvQu+SJKkIRk0xL9XVUcl2Q3Yqapua7MoSZK0dYNOp9+QZBnwBOD2FuuRJEkDGjTEDwL+nd60+g1J/iHJEe2VJUmStmagEK+qn1fVxVX1Z8ChwB7A51utTJIkzWjQPXGSPCnJe4ErgV3pXYZVkiQNyUAntiW5Efg6cDFwalV58xNJkoZs0LPTH1VVt7ZaiSRJmpUZQzzJaVV1NvCmJDV1fVW9bCvPPxp4NzAP+GBVvXUL/Z4FXAocVlXjgxYvSdKObGt74tc2v2cdrEnmAecATwU2AKuTrKiqtVP67Q68HPjKbLchSdKObMYQr6pPNA+vqaqvzXLsw4F1VXU9QJLl9G5nunZKv7OAvwNOneX4kiTt0AY9O/3tSa5NclaSQwZ8zn7A+r72hmbZ3ZI8FlhQVf8200BJliYZTzI+MTEx4OYlSRptg35P/MnAk4EJ4P1Jrrmv9xNPshPwDuBVA2x/WVUtrqrFY2Nj92WzkiSNjIG/J15Vm6rq74EXA1cBZ2zlKRuBBX3t+c2ySbsDhwCfa77C9gRgRZLFg9YkSdKObKAQT/KIJG9Icg3wHuBL9EJ5JquBRUkOSLILsARYMbmyqm6pqr2ramFVLQSuAI717HRJkgYz6PfEzwWWA0+vqh8M8oSq2pzkZGAVva+YnVtVa5KcCYxX1YqZR5AkSTPZaog3XxW7oarePdvBq2olsHLKsmmn4avqyNmOL0nSjmyr0+lVdRewoJkSlyRJc8Sg0+k3AJcnWQHcfd30qnpHK1VJkqStGjTEv9v87ETvrHJJkjRkA4V4Vf1t24VIkqTZGfRWpJ8FprsByh9t84okSdJABp1Of3Xf412BZwGbt305kiRpUINOp185ZdHlSb7aQj2SJGlAg06nP6ivuROwGNizlYokSdJABp1Ov5JfHxPfDNwIvLCNgiRJ0mBmDPEkhwHrq+qApv0CesfDb+Q37wsuSZK2o61dse39wJ0ASZ4IvAU4H7gFWNZuaZIkaSZbm06fV1U/bh7/BbCsqj4GfCzJVe2WJkmSZrK1PfF5SSaD/inAZ/rWDXo8XZIktWBrQfxR4PNJbgZ+AfwHQJKH0ptSlyRJQzJjiFfVm5J8GtgX+FRVTZ6hvhNwStvFSZKkLdvqlHhVXTHNsm+3U44kSRrUVu8nLkmS5iZDXJKkjmo1xJMcneS6JOuSnD7N+hcnuSbJVUm+mOTgNuuRJGmUtBbiSeYB5wDHAAcDJ0wT0hdW1e9W1WOAs4F3tFWPJEmjps098cOBdVV1fVXdCSwHjuvvUFW39jV3Y5p7lkuSpOm1ecGW/YD1fe0NwOOndkryUuCVwC7AH003UJKlwFKA/ffff5sXKklSFw39xLaqOqeqDgT+Gnj9Fvosq6rFVbV4bGxs+xYoSdIc1WaIbwQW9LXnN8u2ZDnwpy3WI0nSSGkzxFcDi5IckGQXYAmwor9DkkV9zT8GvtNiPZIkjZTWjolX1eYkJwOrgHnAuVW1JsmZwHhVrQBOTnIU8EvgJ8AL2qpHkqRR0+qdyKpqJbByyrIz+h6/vM3tS5I0yoZ+YpskSbp3DHFJkjrKEJckqaMMcUmSOsoQlySpowxxSZI6yhCXJKmjDHFJkjrKEJckqaMMcUmSOsoQlySpowxxSZI6yhCXJKmjDHFJkjqq1VuRSv1OO+00Nm3axD777MPZZ5897HIkqfMMcW03mzZtYuPGjcMuQ5JGhtPpkiR1VKshnuToJNclWZfk9GnWvzLJ2iRXJ/l0koe0WY8kSaOktRBPMg84BzgGOBg4IcnBU7p9HVhcVY8CLgU8UCpJ0oDa3BM/HFhXVddX1Z3AcuC4/g5V9dmq+nnTvAKY32I9kiSNlDZDfD9gfV97Q7NsS14IfHK6FUmWJhlPMj4xMbENS5QkqbvmxIltSZ4LLAbeNt36qlpWVYuravHY2Nj2LU6SpDmqza+YbQQW9LXnN8vuIclRwOuAJ1XVHS3WI0nSSGkzxFcDi5IcQC+8lwAn9ndIcijwfuDoqvphi7Vs1as++aFhbn6HcPPPb7v7t3/v9rz9mOcPuwRJ20lr0+lVtRk4GVgFXAtcXFVrkpyZ5Nim29uABwCXJLkqyYq26pEkadS0esW2qloJrJyy7Iy+x0e1uX1JkkbZnDixTZIkzZ4hLklSRxnikiR1lCEuSVJHGeKSJHWUIS5JUkcZ4pIkdVSr3xOX+u2yx273+C1Jum8McW03Bx7/tGGXIEkjxel0SZI6yhCXJKmjDHFJkjrKEJckqaMMcUmSOsoQlySpowxxSZI6yhCXJKmjWg3xJEcnuS7JuiSnT7P+iUm+lmRzkme3WYskSaOmtRBPMg84BzgGOBg4IcnBU7p9HzgJuLCtOiRJGlVtXnb1cGBdVV0PkGQ5cBywdrJDVd3YrPtVi3VIkjSS2pxO3w9Y39fe0CybtSRLk4wnGZ+YmNgmxUmS1HWdOLGtqpZV1eKqWjw2NjbsciRJmhPaDPGNwIK+9vxmmSRJ2gbaDPHVwKIkByTZBVgCrGhxe5Ik7VBaC/Gq2gycDKwCrgUurqo1Sc5McixAksOSbACOB96fZE1b9UiSNGraPDudqloJrJyy7Iy+x6vpTbNLkqRZ6sSJbZIk6TcZ4pIkdZQhLklSRxnikiR1lCEuSVJHGeKSJHWUIS5JUkcZ4pIkdZQhLklSRxnikiR1lCEuSVJHGeKSJHWUIS5JUkcZ4pIkdZQhLklSRxnikiR1lCEuSVJHGeKSJHVUqyGe5Ogk1yVZl+T0adb/VpKLmvVfSbKwzXokSRolrYV4knnAOcAxwMHACUkOntLthcBPquqhwDuBv2urHkmSRk2be+KHA+uq6vqquhNYDhw3pc9xwPnN40uBpyRJizVJkjQydm5x7P2A9X3tDcDjt9SnqjYnuQX4L8DN/Z2SLAWWNs3bk1zXSsXaHvZmyn9fbVvv4AXDLkFzk++9tv3V29oa+SFbWtFmiG8zVbUMWDbsOnTfJRmvqsXDrkPa0fjeG01tTqdvBBb0tec3y6btk2RnYE/gRy3WJEnSyGgzxFcDi5IckGQXYAmwYkqfFXD33N+zgc9UVbVYkyRJI6O16fTmGPfJwCpgHnBuVa1JciYwXlUrgH8CLkiyDvgxvaDXaPOwiDQcvvdGUNzxlSSpm7ximyRJHWWIS5LUUYa4hibJkUn+ddh1SF2Q5GVJrk3ykZbGf0OSV7cxttrTie+JS5L4K+Coqtow7EI0d7gnrvskycIk30pyXpJvJ/lIkqOSXJ7kO0kOb36+nOTrSb6U5OHTjLNbknOTfLXpN/USvdIOK8n7gN8BPpnkddO9V5KclOTjSS5LcmOSk5O8sulzRZIHNf1elGR1km8k+ViS+0+zvQOT/N8kVyb5jyQHbd9XrEEZ4toWHgq8HTio+TkROAJ4NfBa4FvAH1bVocAZwJunGeN19K4TcDjwZOBtSXbbDrVLc15VvRj4Ab33xm5s+b1yCPBnwGHAm4CfN++7LwPPb/r8S1UdVlWPBq6ldyOqqZYBp1TV4+i9j9/bzivTfeV0uraFG6rqGoAka4BPV1UluQZYSO9KfOcnWQQUcL9pxngacGzfMbldgf3p/SMj6de29F4B+GxV3Qbc1tyL4hPN8muARzWPD0nyRmAv4AH0ruVxtyQPAH4fuKTvflS/1cYL0X1niGtbuKPv8a/62r+i9//YWfT+cflvzT3jPzfNGAGeVVXe3Eaa2bTvlSSPZ+vvRYDzgD+tqm8kOQk4csr4OwE/rarHbNuy1Qan07U97Mmvr5t/0hb6rAJOmbwVbZJDt0NdUhfd1/fK7sBNSe4HPGfqyqq6FbghyfHN+Eny6PtYs1piiGt7OBt4S5Kvs+XZn7PoTbNf3UzJn7W9ipM65r6+V/4G+ApwOb3zVabzHOCFSb4BrAE80XSO8rKrkiR1lHvikiR1lCEuSVJHGeKSJHWUIS5JUkcZ4pIkdZQhLuluzXW51yS5OslVzQVEJM1RXrFNEgBJfg/4E+CxVXVHkr2BXYZclqQZuCcuadK+wM1VdQdAVd1cVT9I8rgkn2/uaLUqyb5J9kxy3eQd6ZJ8NMmLhlq9tAPyYi+SgLtvfPFF4P7AvwMXAV8CPg8cV1UTSf4CeHpV/Y8kTwXOBN4NnFRVRw+pdGmH5XS6JACq6vYkjwP+kN4tLi8C3kjv9paXNZfqngfc1PS/rLm+9jmA19aWhsA9cUnTSvJs4KXArlX1e9Os34neXvpC4BmTt6OVtP14TFwSAEke3tzzfdJj6N3Pfaw56Y0k90vyyGb9K5r1JwL/3NwVS9J25J64JACaqfT3AHsBm4F1wFJgPvD39G4puzPwLuALwMeBw6vqtiTvAG6rqv81jNqlHZUhLklSRzmdLklSRxnikiR1lCEuSVJHGeKSJHWUIS5JUkcZ4pIkdZQhLklSR/1/cCdgdb8kKT8AAAAASUVORK5CYII=\n",
            "text/plain": [
              "<Figure size 576x288 with 1 Axes>"
            ]
          },
          "metadata": {
            "tags": [],
            "needs_background": "light"
          }
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "NTaDrMZ2m_PS",
        "outputId": "02ea948b-86fc-4b39-f2c1-2eb9d59d7bb5",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 350
        }
      },
      "source": [
        "# Interestingly, woman survive pretty well along all age classes, while men mostly survive when they are young\n",
        "plt.figure(figsize=(8,5))\n",
        "ax = sns.barplot(x=\"Sex\", y = \"Survived\", hue = \"Age_binned\", data = engineered, palette = 'Set2')\n",
        "plt.legend(loc= \"upper left\")\n",
        "ax.set(title = \"Survival Rate by Gender and Age Group\")\n",
        "plt.show()"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<Figure size 576x360 with 1 Axes>"
            ]
          },
          "metadata": {
            "tags": [],
            "needs_background": "light"
          }
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "_-kvyXKdnKI1"
      },
      "source": [
        "#convert Sex to a numerical feature\n",
        "engineered[\"Sex\"] = cleaned[\"Sex\"].apply(lambda x: 1 if x ==\"male\" else 0)"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "ha9lNdLJoUqb"
      },
      "source": [
        "####**3.3 DISCRETE FEATURES**"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "SFLRDyi8oUxt"
      },
      "source": [
        "**3.3.1 Class**"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "Admi6L6tpVDH",
        "outputId": "4427bae6-fdcb-4859-9500-eec388f2b43b",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 144
        }
      },
      "source": [
        "# Relatively much more passengers survived in first class\n",
        "Class = pd.pivot_table(cleaned, index = \"Survived\", columns = \"Pclass\", values = \"PassengerId\", aggfunc= \"count\")\n",
        "Class.head()"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/html": [
              "<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>Pclass</th>\n",
              "      <th>1</th>\n",
              "      <th>2</th>\n",
              "      <th>3</th>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>Survived</th>\n",
              "      <th></th>\n",
              "      <th></th>\n",
              "      <th></th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>0.0</th>\n",
              "      <td>80</td>\n",
              "      <td>97</td>\n",
              "      <td>372</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1.0</th>\n",
              "      <td>136</td>\n",
              "      <td>87</td>\n",
              "      <td>119</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>"
            ],
            "text/plain": [
              "Pclass      1   2    3\n",
              "Survived              \n",
              "0.0        80  97  372\n",
              "1.0       136  87  119"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 120
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "lBzyXUM7oyCB",
        "outputId": "a6e86b7c-47f4-4ec2-c090-a59034f3a6e5",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 350
        }
      },
      "source": [
        "# We can confirm the lower survival of third class passengers with a barplot\n",
        "plt.figure(figsize=(8,5))\n",
        "ax = sns.barplot(x=\"Pclass\", y = \"Survived\",  data = engineered, palette = 'Set2')\n",
        "ax.set(title = \"Survival Rate by Class\")\n",
        "plt.show()"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<Figure size 576x360 with 1 Axes>"
            ]
          },
          "metadata": {
            "tags": [],
            "needs_background": "light"
          }
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "F6bETwSxqL8Q"
      },
      "source": [
        "**3.3.2 SIBSP & PARCH**"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "JwXl_JZsqUyR"
      },
      "source": [
        "SibSp indicates the number of siblings and spouses aboard the Titanic. Parch indicates the number of children or parents aboard. In earlier versions of the notebook I used the sum of these two features to indicate the group size a person was travelling with. However, this feature turned out to be a poor representation of the actual group size, because it doesn't include cousins, nannies, friends etc. Further on in this notebook we will create a better feature to figure out who is travelling with whom. "
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "H2gp5ohxtGok"
      },
      "source": [
        "####**3.4 NOMINAL FEATURES**"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "2jp7DPGlnBMW"
      },
      "source": [
        "**3.4.1 TITLE**"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "pZdcZ-MQjZ5c",
        "outputId": "a4031970-cc48-4cf7-9556-b2399a4c15f5",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 144
        }
      },
      "source": [
        "#It looks like a lot of people with the title \"Mr\" died while a lot of \"Mrs\", \"Miss\" and \"Master\" died.\n",
        "title = pd.pivot_table(cleaned, index = \"Survived\", columns = \"Title\", values = \"PassengerId\", aggfunc= \"count\")\n",
        "title"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/html": [
              "<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>Title</th>\n",
              "      <th>Capt</th>\n",
              "      <th>Col</th>\n",
              "      <th>Countess</th>\n",
              "      <th>Don</th>\n",
              "      <th>Dr</th>\n",
              "      <th>Jonkheer</th>\n",
              "      <th>Lady</th>\n",
              "      <th>Major</th>\n",
              "      <th>Master</th>\n",
              "      <th>Miss</th>\n",
              "      <th>Mlle</th>\n",
              "      <th>Mme</th>\n",
              "      <th>Mr</th>\n",
              "      <th>Mrs</th>\n",
              "      <th>Ms</th>\n",
              "      <th>Rev</th>\n",
              "      <th>Sir</th>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>Survived</th>\n",
              "      <th></th>\n",
              "      <th></th>\n",
              "      <th></th>\n",
              "      <th></th>\n",
              "      <th></th>\n",
              "      <th></th>\n",
              "      <th></th>\n",
              "      <th></th>\n",
              "      <th></th>\n",
              "      <th></th>\n",
              "      <th></th>\n",
              "      <th></th>\n",
              "      <th></th>\n",
              "      <th></th>\n",
              "      <th></th>\n",
              "      <th></th>\n",
              "      <th></th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>0.0</th>\n",
              "      <td>1.0</td>\n",
              "      <td>1.0</td>\n",
              "      <td>NaN</td>\n",
              "      <td>1.0</td>\n",
              "      <td>4.0</td>\n",
              "      <td>1.0</td>\n",
              "      <td>NaN</td>\n",
              "      <td>1.0</td>\n",
              "      <td>17.0</td>\n",
              "      <td>55.0</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>436.0</td>\n",
              "      <td>26.0</td>\n",
              "      <td>NaN</td>\n",
              "      <td>6.0</td>\n",
              "      <td>NaN</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1.0</th>\n",
              "      <td>NaN</td>\n",
              "      <td>1.0</td>\n",
              "      <td>1.0</td>\n",
              "      <td>NaN</td>\n",
              "      <td>3.0</td>\n",
              "      <td>NaN</td>\n",
              "      <td>1.0</td>\n",
              "      <td>1.0</td>\n",
              "      <td>23.0</td>\n",
              "      <td>127.0</td>\n",
              "      <td>2.0</td>\n",
              "      <td>1.0</td>\n",
              "      <td>81.0</td>\n",
              "      <td>99.0</td>\n",
              "      <td>1.0</td>\n",
              "      <td>NaN</td>\n",
              "      <td>1.0</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>"
            ],
            "text/plain": [
              "Title     Capt  Col  Countess  Don   Dr  ...     Mr   Mrs   Ms  Rev  Sir\n",
              "Survived                                 ...                            \n",
              "0.0        1.0  1.0       NaN  1.0  4.0  ...  436.0  26.0  NaN  6.0  NaN\n",
              "1.0        NaN  1.0       1.0  NaN  3.0  ...   81.0  99.0  1.0  NaN  1.0\n",
              "\n",
              "[2 rows x 17 columns]"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 139
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "NnhPxVxFumTQ"
      },
      "source": [
        "It is pretty important that people (especially men) in the most common title groups have the correct title. I noticed that the oldest boy with the title \"Master\" was 40, and the youngest \"Mr\" was 1. We will have to correct these titles."
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "kJ7bEreb68ur",
        "outputId": "385dad7d-7322-4ac8-91a3-dad844830b57",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 645
        }
      },
      "source": [
        "#check the median and mean Fare per class and title (only of training data to avoid leakage)\n",
        "mini = engineered.groupby([ \"Title\"])[\"Age\"].min()\n",
        "maxi = engineered.groupby([ \"Title\"])[\"Age\"].max()\n",
        "mean = engineered.groupby([ \"Title\"])[\"Age\"].mean()\n",
        "age = pd.concat([mini, mean, maxi], axis=1, keys = [\"min\", \"avg\", \"max\"])\n",
        "age"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/html": [
              "<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>min</th>\n",
              "      <th>avg</th>\n",
              "      <th>max</th>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>Title</th>\n",
              "      <th></th>\n",
              "      <th></th>\n",
              "      <th></th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>Capt</th>\n",
              "      <td>70</td>\n",
              "      <td>70.000000</td>\n",
              "      <td>70</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>Col</th>\n",
              "      <td>47</td>\n",
              "      <td>54.000000</td>\n",
              "      <td>60</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>Countess</th>\n",
              "      <td>33</td>\n",
              "      <td>33.000000</td>\n",
              "      <td>33</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>Don</th>\n",
              "      <td>40</td>\n",
              "      <td>40.000000</td>\n",
              "      <td>40</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>Dona</th>\n",
              "      <td>39</td>\n",
              "      <td>39.000000</td>\n",
              "      <td>39</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>Dr</th>\n",
              "      <td>23</td>\n",
              "      <td>43.875000</td>\n",
              "      <td>54</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>Jonkheer</th>\n",
              "      <td>38</td>\n",
              "      <td>38.000000</td>\n",
              "      <td>38</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>Lady</th>\n",
              "      <td>48</td>\n",
              "      <td>48.000000</td>\n",
              "      <td>48</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>Major</th>\n",
              "      <td>45</td>\n",
              "      <td>48.500000</td>\n",
              "      <td>52</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>Master</th>\n",
              "      <td>0</td>\n",
              "      <td>6.901639</td>\n",
              "      <td>40</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>Miss</th>\n",
              "      <td>0</td>\n",
              "      <td>21.603846</td>\n",
              "      <td>63</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>Mlle</th>\n",
              "      <td>24</td>\n",
              "      <td>24.000000</td>\n",
              "      <td>24</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>Mme</th>\n",
              "      <td>24</td>\n",
              "      <td>24.000000</td>\n",
              "      <td>24</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>Mr</th>\n",
              "      <td>1</td>\n",
              "      <td>31.365918</td>\n",
              "      <td>80</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>Mrs</th>\n",
              "      <td>4</td>\n",
              "      <td>36.385787</td>\n",
              "      <td>76</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>Ms</th>\n",
              "      <td>28</td>\n",
              "      <td>29.000000</td>\n",
              "      <td>30</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>Rev</th>\n",
              "      <td>27</td>\n",
              "      <td>41.250000</td>\n",
              "      <td>57</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>Sir</th>\n",
              "      <td>49</td>\n",
              "      <td>49.000000</td>\n",
              "      <td>49</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>"
            ],
            "text/plain": [
              "          min        avg  max\n",
              "Title                        \n",
              "Capt       70  70.000000   70\n",
              "Col        47  54.000000   60\n",
              "Countess   33  33.000000   33\n",
              "Don        40  40.000000   40\n",
              "Dona       39  39.000000   39\n",
              "Dr         23  43.875000   54\n",
              "Jonkheer   38  38.000000   38\n",
              "Lady       48  48.000000   48\n",
              "Major      45  48.500000   52\n",
              "Master      0   6.901639   40\n",
              "Miss        0  21.603846   63\n",
              "Mlle       24  24.000000   24\n",
              "Mme        24  24.000000   24\n",
              "Mr          1  31.365918   80\n",
              "Mrs         4  36.385787   76\n",
              "Ms         28  29.000000   30\n",
              "Rev        27  41.250000   57\n",
              "Sir        49  49.000000   49"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 145
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "e93601RF68yE"
      },
      "source": [
        "#Correct the title of adult men that where labeled as  \"Master\"\n",
        "mask = (engineered[\"Title\"] == \"Master\") & (engineered[\"Age\"]>17)\n",
        "wrong_title = engineered.loc[mask]\n",
        "engineered.loc[mask, \"Title\"] = \"Mr\""
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "RrHOo7NMNGJI"
      },
      "source": [
        "#Correct the title of boys that where labeled as \"Mr\"\n",
        "mask = (engineered[\"Title\"] == \"Mr\") & (engineered[\"Age\"]<18)\n",
        "engineered.loc[mask, \"Title\"] = \"Master\""
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "zHiB2hbUng73"
      },
      "source": [
        "# replacing some titles and grouping them in main categories\n",
        "engineered['Title'] = engineered['Title'].replace(['Lady', 'Countess','Capt', 'Col','Don', 'Dr',\\\n",
        "                                            'Major', 'Rev', 'Sir', 'Jonkheer', 'Dona', 'Ms'], 'Other')\n",
        "engineered['Title'] = engineered['Title'].replace('Mlle', 'Miss')\n",
        "engineered['Title'] = engineered['Title'].replace('Mme', 'Mrs')\n"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "G8TeFb8pnyfg"
      },
      "source": [
        "**3.4.2 SURNAME**"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "TKgSFn0wvU4E"
      },
      "source": [
        "I spend a lot of time on the Surname feature. My hypothesis was that non-native speakers were more likely to die, because they wouldn't be able to understand the crews instructions. I compiled a list of thousands of common English and Irish surnames and found..... NOTHING!!! \n",
        "\n",
        "\n",
        "After doing some research, I found that a lot of Irish people travelled in third class, and perished. Many American passengers lived, while many British passengers died. "
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "r1e6x6Rqng-v"
      },
      "source": [
        "# we can easily get the surnames by splitting the dataset at the comma and grabbing the first value. I won't use this feature here.\n",
        "engineered[\"Surname\"]= cleaned[\"Name\"].str.split(\",\").str.get(0)"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "gxux-xYQweTe"
      },
      "source": [
        "**3.4.3 TICKET**"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "6KxQltluwp1d"
      },
      "source": [
        "I thought it would be interesting to figure out who was travelling with whom. \n",
        "\n",
        "\n",
        "I found out that there is no easy way to identify groups. If you look at surnames, you will misidentify people who share the same name, but don't travel together. You will also miss people who travel together but have different surnames. If you look at the ticket number, you will include people who travel together but have different surnames, but miss people whose ticket is one number off. \n",
        "\n",
        "You have no way of knowing for sure who is travelling together. For example, The Anderssons travel together in a group of 3 on the same ticket number. Lundahl travels alone, but his ticket number is one number of from the Anderssons. Is this a coincidence? Or is Lundahl a friend or cousin of the Anderssons? \n",
        "\n",
        "After checking a lot of tickets and surnames, I felt that ticket number would be the best way to identify people travelling together.  That is, until I found a better way of identifying groups.\n",
        "\n"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "9Q0ko3j0wVnN"
      },
      "source": [
        "#This is how I extracted ticket numbers. I won't actually use this feature here.\n",
        "#extract TicketNumber: if Ticket number is \"LINE\", replace by 0\n",
        "engineered[\"TicketNr\"] = engineered[\"Ticket\"].str.split(\" \").str.get(-1)\n",
        "engineered[\"TicketNr\"]  = engineered[\"TicketNr\"] .replace(\"LINE\",0)\n",
        "engineered.TicketNr[engineered.TicketNr == 0] = np.nan"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "oGVqLYnv2diq"
      },
      "source": [
        "**3.4.4 GROUPS**"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "0i_2VsB32jOg"
      },
      "source": [
        "Before I was able to properly identify groups, I scored a 78.9% accuracy on Kaggle, a top 12% score. Not bad at all! After researching on how I could improve my score, I came across the kernel of [David Silva](https://www.kaggle.com/dr1t10/surviving-the-titanic-step-by-step-with-groups). Based on several other high scoring kernels, he identifies groups based on a combination of surname, ticket, port and class. I used his code to build a few new features."
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "qLU3xR82wVsP",
        "outputId": "4a3e1a03-95eb-4ccb-a336-b5d447c632c1",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 311
        }
      },
      "source": [
        "#define groups using surname, ticket (minus the last digit), class and port. \n",
        "surname = engineered['Name'].apply(lambda x: x[:x.find(',')])\n",
        "ticket = engineered['Ticket'].apply(lambda x: x[:-1])\n",
        "\n",
        "engineered['SPTE'] = (surname.astype(str) + '-' + engineered['Pclass'].astype(str) + '-'\n",
        "           + ticket.astype(str) + '-' + engineered['Embarked'].astype(str))\n",
        "\n",
        "spte_count = engineered['SPTE'].value_counts(sort=False)\n",
        "\n",
        "def spte_group_labeller(group):\n",
        "    group_elements = engineered.loc[engineered['SPTE'] == group, 'PassengerId']\n",
        "    if len(group_elements) == 1:\n",
        "        return 0\n",
        "    else:\n",
        "        return group_elements.min()\n",
        "\n",
        "engineered['GroupId'] = engineered['SPTE'].apply(spte_group_labeller)\n",
        "#engineered.drop(columns='SPTE', inplace=True)\n",
        "engineered.tail()"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/html": [
              "<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>PassengerId</th>\n",
              "      <th>Survived</th>\n",
              "      <th>Pclass</th>\n",
              "      <th>Name</th>\n",
              "      <th>Sex</th>\n",
              "      <th>Age</th>\n",
              "      <th>SibSp</th>\n",
              "      <th>Parch</th>\n",
              "      <th>Ticket</th>\n",
              "      <th>Fare</th>\n",
              "      <th>Cabin</th>\n",
              "      <th>Embarked</th>\n",
              "      <th>Title</th>\n",
              "      <th>Age_binned</th>\n",
              "      <th>Fare_binned</th>\n",
              "      <th>Surname</th>\n",
              "      <th>TicketNr</th>\n",
              "      <th>GroupId</th>\n",
              "      <th>GroupSize</th>\n",
              "      <th>SPTE</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>413</th>\n",
              "      <td>1305</td>\n",
              "      <td>NaN</td>\n",
              "      <td>3</td>\n",
              "      <td>Spector, Mr. Woolf</td>\n",
              "      <td>1</td>\n",
              "      <td>31</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>A.5. 3236</td>\n",
              "      <td>8.0500</td>\n",
              "      <td>U</td>\n",
              "      <td>S</td>\n",
              "      <td>Mr</td>\n",
              "      <td>Young Adult</td>\n",
              "      <td>Very Low</td>\n",
              "      <td>Spector</td>\n",
              "      <td>3236</td>\n",
              "      <td>0</td>\n",
              "      <td>748</td>\n",
              "      <td>Spector-3-A.5. 323-S</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>414</th>\n",
              "      <td>1306</td>\n",
              "      <td>NaN</td>\n",
              "      <td>1</td>\n",
              "      <td>Oliva y Ocana, Dona. Fermina</td>\n",
              "      <td>0</td>\n",
              "      <td>39</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>PC 17758</td>\n",
              "      <td>108.9000</td>\n",
              "      <td>C105</td>\n",
              "      <td>C</td>\n",
              "      <td>Other</td>\n",
              "      <td>Adult</td>\n",
              "      <td>High</td>\n",
              "      <td>Oliva y Ocana</td>\n",
              "      <td>17758</td>\n",
              "      <td>0</td>\n",
              "      <td>3</td>\n",
              "      <td>Oliva y Ocana-1-PC 1775-C</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>415</th>\n",
              "      <td>1307</td>\n",
              "      <td>NaN</td>\n",
              "      <td>3</td>\n",
              "      <td>Saether, Mr. Simon Sivertsen</td>\n",
              "      <td>1</td>\n",
              "      <td>38</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>SOTON/O.Q. 3101262</td>\n",
              "      <td>7.2500</td>\n",
              "      <td>U</td>\n",
              "      <td>S</td>\n",
              "      <td>Mr</td>\n",
              "      <td>Adult</td>\n",
              "      <td>Very Low</td>\n",
              "      <td>Saether</td>\n",
              "      <td>3101262</td>\n",
              "      <td>0</td>\n",
              "      <td>748</td>\n",
              "      <td>Saether-3-SOTON/O.Q. 310126-S</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>416</th>\n",
              "      <td>1308</td>\n",
              "      <td>NaN</td>\n",
              "      <td>3</td>\n",
              "      <td>Ware, Mr. Frederick</td>\n",
              "      <td>1</td>\n",
              "      <td>32</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>359309</td>\n",
              "      <td>8.0500</td>\n",
              "      <td>U</td>\n",
              "      <td>S</td>\n",
              "      <td>Mr</td>\n",
              "      <td>Young Adult</td>\n",
              "      <td>Very Low</td>\n",
              "      <td>Ware</td>\n",
              "      <td>359309</td>\n",
              "      <td>0</td>\n",
              "      <td>748</td>\n",
              "      <td>Ware-3-35930-S</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>417</th>\n",
              "      <td>1309</td>\n",
              "      <td>NaN</td>\n",
              "      <td>3</td>\n",
              "      <td>Peter, Master. Michael J</td>\n",
              "      <td>1</td>\n",
              "      <td>24</td>\n",
              "      <td>1</td>\n",
              "      <td>1</td>\n",
              "      <td>2668</td>\n",
              "      <td>22.3583</td>\n",
              "      <td>U</td>\n",
              "      <td>C</td>\n",
              "      <td>Master</td>\n",
              "      <td>Young Adult</td>\n",
              "      <td>Average</td>\n",
              "      <td>Peter</td>\n",
              "      <td>2668</td>\n",
              "      <td>129</td>\n",
              "      <td>3</td>\n",
              "      <td>Peter-3-266-C</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>"
            ],
            "text/plain": [
              "     PassengerId  Survived  ...  GroupSize                           SPTE\n",
              "413         1305       NaN  ...        748           Spector-3-A.5. 323-S\n",
              "414         1306       NaN  ...          3      Oliva y Ocana-1-PC 1775-C\n",
              "415         1307       NaN  ...        748  Saether-3-SOTON/O.Q. 310126-S\n",
              "416         1308       NaN  ...        748                 Ware-3-35930-S\n",
              "417         1309       NaN  ...          3                  Peter-3-266-C\n",
              "\n",
              "[5 rows x 20 columns]"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 193
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "kzmuP3n54BGN",
        "outputId": "fd87600f-6c67-4b94-db8a-c38c35eacec4",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 35
        }
      },
      "source": [
        "#we can identify 199 different groups\n",
        "engineered['GroupId'].nunique()"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "199"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 194
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "U5G_8fmu6iMC"
      },
      "source": [
        "**3.4.5 GROUP SIZE**"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "Y2PACDiywVx7"
      },
      "source": [
        "#We can identify the groupsize of each GroupId by counting the passengerId's for each group\n",
        "engineered[\"GroupSize\"] = engineered.groupby(\"GroupId\")[\"PassengerId\"].transform('count')"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "XBKtDhWG69zk"
      },
      "source": [
        "#Most first class passengers boarded in Southampton\n",
        "ax = sns.countplot(x=\"Pclass\", hue=\"Embarked\", data=raw, palette = 'Set2')\n",
        "ax.set(title = \"Embarkation port by Class\")\n",
        "plt.show()"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "jHqAnKrS6oD-",
        "outputId": "f505dba7-f1a8-49e6-ce20-d8c015d1ac6f",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 295
        }
      },
      "source": [
        "#People who travelled alone (code 748) or in groups of 5 or more had a much lower chance of survival\n",
        "ax = sns.barplot(x=\"GroupSize\", y = \"Survived\", data = engineered, palette = \"Set2\")\n",
        "ax.set(title = \"Survival Rate by Group Size\")\n",
        "plt.show()"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<Figure size 432x288 with 1 Axes>"
            ]
          },
          "metadata": {
            "tags": [],
            "needs_background": "light"
          }
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "tUMxDKvJ-SIG"
      },
      "source": [
        "**3.4.6 COMPANY**"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "vjSqEZWi6oIm"
      },
      "source": [
        "#create feature Company\n",
        "engineered[\"Company\"] = engineered[\"GroupSize\"].apply(lambda x: \"Alone\" if x == 794 else \"Pair\" if x == 2 else \"Small Group\" if x <5 else \"Large Group\")"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "Ks1A4xRc-dgA",
        "outputId": "0e845c64-51ff-4fc8-c3b7-5b3813e892a9",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 295
        }
      },
      "source": [
        "#People who travelled in a small group or in a pair had a higher chance of surviving\n",
        "ax =sns.barplot(x=\"Company\", y = \"Survived\", data = engineered, palette = \"Set2\")\n",
        "ax.set(title = \"Survival rate by group size\")\n",
        "plt.show()"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<Figure size 432x288 with 1 Axes>"
            ]
          },
          "metadata": {
            "tags": [],
            "needs_background": "light"
          }
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "nm_nBF6K_P_b"
      },
      "source": [
        "**3.4.7 IN GROUP**"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "Z6gGwep4_VKa"
      },
      "source": [
        "#Is the passenger travelling in a group or not\n",
        "engineered[\"InGroup\"] = engineered[\"GroupId\"].apply(lambda x: 1 if x >0 else 0)"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "HeiwGkXYAT8W",
        "outputId": "1b4d3b06-4abc-4ebe-ee31-0980997fb2fd",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 295
        }
      },
      "source": [
        "#travelling in a group increases the chance of survival in all classes\n",
        "ax = sns.barplot(x=\"InGroup\", y = \"Survived\", hue = \"Pclass\" , data = engineered, palette = \"Set2\")\n",
        "ax.set(title = 'Survival by Travel in Group')\n",
        "plt.show()"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<Figure size 432x288 with 1 Axes>"
            ]
          },
          "metadata": {
            "tags": [],
            "needs_background": "light"
          }
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "qNwUWRKtAmD_"
      },
      "source": [
        "**3.4.8 WOMAND CHILD GROUPS**"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "3mpwE53R6IEp"
      },
      "source": [
        "[Chris Deotte](https://www.kaggle.com/cdeotte/titantic-mega-model-0-84210) noticed how nearly all  80 woman-child-groups either all lived or all died (based on the survival information we know). That is the secret behind the woman-child-group model. Identifying women-child groups is therefore interesting. Credits to [David Silva](https://www.kaggle.com/dr1t10/surviving-the-titanic-step-by-step-with-groups) for the code. Chris and David also experiment with creating features that indicate whether all people in a Woman Child Group live or die. This could be interesting to explore some time in the future."
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "kjizGCvbH4eL"
      },
      "source": [
        "# Create a mask to filter out groups (i.e. people not travelling alone) containing woman and children\n",
        "mask = (engineered[\"Company\"] != \"Alone\") & ((engineered[\"Sex\"]== 0) | (engineered[\"Title\"] == \"Master\"))\n",
        "\n",
        "# Count the number of woman and children (mask) for each GroupId (set GroupId as index); Get rid of groups with only one woman or child\n",
        "WomanChildGroups = engineered.loc[mask, 'GroupId'].value_counts()\n",
        "WomanChildGroups = WomanChildGroups[WomanChildGroups > 1]\n",
        "\n",
        "# Update the mask to discard groups with only one female or boy\n",
        "mask = mask & (engineered['GroupId'].isin(WomanChildGroups.index))\n",
        "\n",
        "# Create the new feature using the updated mask\n",
        "engineered['WomanChildGroup'] = 0\n",
        "engineered.loc[mask, 'WomanChildGroup'] = 1"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "MusU_CJh-dix",
        "outputId": "cc863e0d-f45d-40d5-a6bd-41bf3b3bf520",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 295
        }
      },
      "source": [
        "#people travelling in a woman and child group have a higher chance of survival\n",
        "ax = sns.barplot(x=\"WomanChildGroup\", y = \"Survived\" , data = engineered, palette = \"Set2\")\n",
        "ax.set(title = 'Survival by Woman Child Group')\n",
        "plt.show()"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<Figure size 432x288 with 1 Axes>"
            ]
          },
          "metadata": {
            "tags": [],
            "needs_background": "light"
          }
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "ZM3A3DLICArO"
      },
      "source": [
        "####**3.5 ORDINAL FEATURES**"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "RzTcXnNLs_kY"
      },
      "source": [
        "**3.5.1 DECK**"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "HZsrMX2onAV6"
      },
      "source": [
        "#extract the Deck from the Cabin feature\n",
        "engineered[\"Deck\"] = cleaned[\"Cabin\"].str[0]"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "zJQiID5-CN92",
        "outputId": "f13d436c-9c38-4c4c-9e75-f227a2642a13",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 295
        }
      },
      "source": [
        "#People with unknown Deck ID's tend to die, but there is a lot of variation in this feature.\n",
        "ax = sns.barplot(x=\"Deck\", y = \"Survived\" ,data = engineered, palette = \"Set2\")\n",
        "ax.set(title = 'Survival by Woman Child Group')\n",
        "plt.show()"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<Figure size 432x288 with 1 Axes>"
            ]
          },
          "metadata": {
            "tags": [],
            "needs_background": "light"
          }
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "vEwHSP4ZComV",
        "outputId": "345b5cb4-d75d-4498-97b6-5e1a5eed0970",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 175
        }
      },
      "source": [
        "#Passengers with unknown Deck most often travel in third class. \n",
        "Deck= pd.pivot_table(engineered, index = \"Pclass\", columns = \"Deck\", values = \"PassengerId\", aggfunc= \"count\")\n",
        "Deck"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/html": [
              "<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>Deck</th>\n",
              "      <th>A</th>\n",
              "      <th>B</th>\n",
              "      <th>C</th>\n",
              "      <th>D</th>\n",
              "      <th>E</th>\n",
              "      <th>F</th>\n",
              "      <th>G</th>\n",
              "      <th>T</th>\n",
              "      <th>U</th>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>Pclass</th>\n",
              "      <th></th>\n",
              "      <th></th>\n",
              "      <th></th>\n",
              "      <th></th>\n",
              "      <th></th>\n",
              "      <th></th>\n",
              "      <th></th>\n",
              "      <th></th>\n",
              "      <th></th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>1</th>\n",
              "      <td>22.0</td>\n",
              "      <td>65.0</td>\n",
              "      <td>94.0</td>\n",
              "      <td>40.0</td>\n",
              "      <td>34.0</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>1.0</td>\n",
              "      <td>67.0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2</th>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>6.0</td>\n",
              "      <td>4.0</td>\n",
              "      <td>13.0</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>254.0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>3</th>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>3.0</td>\n",
              "      <td>8.0</td>\n",
              "      <td>5.0</td>\n",
              "      <td>NaN</td>\n",
              "      <td>693.0</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>"
            ],
            "text/plain": [
              "Deck       A     B     C     D     E     F    G    T      U\n",
              "Pclass                                                     \n",
              "1       22.0  65.0  94.0  40.0  34.0   NaN  NaN  1.0   67.0\n",
              "2        NaN   NaN   NaN   6.0   4.0  13.0  NaN  NaN  254.0\n",
              "3        NaN   NaN   NaN   NaN   3.0   8.0  5.0  NaN  693.0"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 221
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "hte8bToDCzh8"
      },
      "source": [
        "I though Deck would be an ordinal feature, but looking at the survival by deck, it isn't really. Deck is extracted from the Cabin feature. Having a cabin close to a staircase would probably have helped. I will therefore use onehot encoding on this feature."
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "p8f4-vciDlgO"
      },
      "source": [
        "####**3.6 SCALING & ENCODING**"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "INDa-haem-LO",
        "outputId": "5c382f9d-3327-48bf-b76a-439d90b7c5cb",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 271
        }
      },
      "source": [
        "engineered.head(2)"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/html": [
              "<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>PassengerId</th>\n",
              "      <th>Survived</th>\n",
              "      <th>Pclass</th>\n",
              "      <th>Name</th>\n",
              "      <th>Sex</th>\n",
              "      <th>Age</th>\n",
              "      <th>SibSp</th>\n",
              "      <th>Parch</th>\n",
              "      <th>Ticket</th>\n",
              "      <th>Fare</th>\n",
              "      <th>Cabin</th>\n",
              "      <th>Embarked</th>\n",
              "      <th>Title</th>\n",
              "      <th>Age_binned</th>\n",
              "      <th>Fare_binned</th>\n",
              "      <th>Surname</th>\n",
              "      <th>TicketNr</th>\n",
              "      <th>GroupId</th>\n",
              "      <th>GroupSize</th>\n",
              "      <th>SPTE</th>\n",
              "      <th>Company</th>\n",
              "      <th>InGroup</th>\n",
              "      <th>WomanChildGroup</th>\n",
              "      <th>Deck</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>0</th>\n",
              "      <td>1</td>\n",
              "      <td>0.0</td>\n",
              "      <td>3</td>\n",
              "      <td>Braund, Mr. Owen Harris</td>\n",
              "      <td>1</td>\n",
              "      <td>22</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>A/5 21171</td>\n",
              "      <td>7.2500</td>\n",
              "      <td>U</td>\n",
              "      <td>S</td>\n",
              "      <td>Mr</td>\n",
              "      <td>Young Adult</td>\n",
              "      <td>Very Low</td>\n",
              "      <td>Braund</td>\n",
              "      <td>21171</td>\n",
              "      <td>0</td>\n",
              "      <td>794</td>\n",
              "      <td>Braund-3-A/5 2117-S</td>\n",
              "      <td>Alone</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>U</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1</th>\n",
              "      <td>2</td>\n",
              "      <td>1.0</td>\n",
              "      <td>1</td>\n",
              "      <td>Cumings, Mrs. John Bradley (Florence Briggs Th...</td>\n",
              "      <td>0</td>\n",
              "      <td>38</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>PC 17599</td>\n",
              "      <td>71.2833</td>\n",
              "      <td>C85</td>\n",
              "      <td>C</td>\n",
              "      <td>Mrs</td>\n",
              "      <td>Adult</td>\n",
              "      <td>High</td>\n",
              "      <td>Cumings</td>\n",
              "      <td>17599</td>\n",
              "      <td>2</td>\n",
              "      <td>2</td>\n",
              "      <td>Cumings-1-PC 1759-C</td>\n",
              "      <td>Pair</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>C</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>"
            ],
            "text/plain": [
              "   PassengerId  Survived  Pclass  ... InGroup  WomanChildGroup  Deck\n",
              "0            1       0.0       3  ...       0                0     U\n",
              "1            2       1.0       1  ...       1                0     C\n",
              "\n",
              "[2 rows x 24 columns]"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 226
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "XXuwvT6gI_m0",
        "outputId": "24aa3980-df9a-4891-f842-b2abf7241bc2",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 206
        }
      },
      "source": [
        "#there are a few features that we can drop\n",
        "final = engineered.drop([\"Name\",\"SibSp\",\"Fare\", \"Parch\", \"Ticket\", \"Cabin\", \"Surname\", \"TicketNr\", \"SPTE\", \"GroupId\", \"GroupSize\"],axis=1)\n",
        "final.head()"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/html": [
              "<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>PassengerId</th>\n",
              "      <th>Survived</th>\n",
              "      <th>Pclass</th>\n",
              "      <th>Sex</th>\n",
              "      <th>Age</th>\n",
              "      <th>Embarked</th>\n",
              "      <th>Title</th>\n",
              "      <th>Age_binned</th>\n",
              "      <th>Fare_binned</th>\n",
              "      <th>Company</th>\n",
              "      <th>InGroup</th>\n",
              "      <th>WomanChildGroup</th>\n",
              "      <th>Deck</th>\n",
              "      <th>FareLog</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>0</th>\n",
              "      <td>1</td>\n",
              "      <td>0.0</td>\n",
              "      <td>3</td>\n",
              "      <td>1</td>\n",
              "      <td>22</td>\n",
              "      <td>S</td>\n",
              "      <td>Mr</td>\n",
              "      <td>Young Adult</td>\n",
              "      <td>Very Low</td>\n",
              "      <td>Alone</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>U</td>\n",
              "      <td>2.110213</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1</th>\n",
              "      <td>2</td>\n",
              "      <td>1.0</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>38</td>\n",
              "      <td>C</td>\n",
              "      <td>Mrs</td>\n",
              "      <td>Adult</td>\n",
              "      <td>High</td>\n",
              "      <td>Pair</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>C</td>\n",
              "      <td>4.280593</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2</th>\n",
              "      <td>3</td>\n",
              "      <td>1.0</td>\n",
              "      <td>3</td>\n",
              "      <td>0</td>\n",
              "      <td>26</td>\n",
              "      <td>S</td>\n",
              "      <td>Miss</td>\n",
              "      <td>Young Adult</td>\n",
              "      <td>Very Low</td>\n",
              "      <td>Alone</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>U</td>\n",
              "      <td>2.188856</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>3</th>\n",
              "      <td>4</td>\n",
              "      <td>1.0</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>35</td>\n",
              "      <td>S</td>\n",
              "      <td>Mrs</td>\n",
              "      <td>Young Adult</td>\n",
              "      <td>Above Average</td>\n",
              "      <td>Pair</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>C</td>\n",
              "      <td>3.990834</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>4</th>\n",
              "      <td>5</td>\n",
              "      <td>0.0</td>\n",
              "      <td>3</td>\n",
              "      <td>1</td>\n",
              "      <td>35</td>\n",
              "      <td>S</td>\n",
              "      <td>Mr</td>\n",
              "      <td>Young Adult</td>\n",
              "      <td>Very Low</td>\n",
              "      <td>Alone</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>U</td>\n",
              "      <td>2.202765</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>"
            ],
            "text/plain": [
              "   PassengerId  Survived  Pclass  Sex  ...  InGroup WomanChildGroup Deck   FareLog\n",
              "0            1       0.0       3    1  ...        0               0    U  2.110213\n",
              "1            2       1.0       1    0  ...        1               0    C  4.280593\n",
              "2            3       1.0       3    0  ...        0               0    U  2.188856\n",
              "3            4       1.0       1    0  ...        1               0    C  3.990834\n",
              "4            5       0.0       3    1  ...        0               0    U  2.202765\n",
              "\n",
              "[5 rows x 14 columns]"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 232
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "SHSoWHe2EJxL"
      },
      "source": [
        "**3.6.1 SCALING**"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "WscWODGfJW5_"
      },
      "source": [
        "#split data in train and test\n",
        "final_train = final[final[\"PassengerId\"].isin(train[\"PassengerId\"])]\n",
        "final_test =  final[final[\"PassengerId\"].isin(test[\"PassengerId\"])]"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "ypy5CIdGGTdM"
      },
      "source": [
        "# Instantiate StandardScaler\n",
        "SS_scaler = StandardScaler()\n",
        "\n",
        "#Age\n",
        "SS_scaler.fit(final_train[['Age']])\n",
        "final_train['Age_S'] = SS_scaler.transform(final_train[['Age']])\n",
        "final_test['Age_S'] = SS_scaler.transform(final_test[['Age']])\n",
        "\n",
        "#Fare\n",
        "#fit and transform FareLog\n",
        "SS_scaler.fit(final_train[['FareLog']])\n",
        "final_train['FareLog_S'] = SS_scaler.transform(final_train[['FareLog']])\n",
        "final_test['FareLog_S'] = SS_scaler.transform(final_test[['FareLog']])\n",
        "\n",
        "#merg train and test together \n",
        "final = pd.concat([final_train, final_test])\n",
        "final = final.drop([\"Age\", \"FareLog\"], axis=1)"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "axTlZ8i8EvtW"
      },
      "source": [
        "**3.6.2 ONE HOT ENCODING**"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "gY1t5z-UKEkw"
      },
      "source": [
        "#one hot encode Deck\n",
        "deck = pd.get_dummies(final['Deck'], prefix=\"Deck\")\n",
        "final = pd.concat([final, deck], axis=1)\n",
        "\n",
        "#one hot encode Embarked\n",
        "embarked = pd.get_dummies(final['Embarked'], prefix=\"Port\")\n",
        "final = pd.concat([final, embarked], axis=1)\n",
        "\n",
        "#one hot encode Title\n",
        "title = pd.get_dummies(final['Title'], prefix=\"Title\")\n",
        "final = pd.concat([final, title], axis=1)\n",
        "\n",
        "#one hot encode Age binned\n",
        "title = pd.get_dummies(final['Age_binned'], prefix=\"Age\")\n",
        "final = pd.concat([final, title], axis=1)\n",
        "\n",
        "#one hot encode Fare binned\n",
        "title = pd.get_dummies(final['Fare_binned'], prefix=\"Fare\")\n",
        "final = pd.concat([final, title], axis=1)\n",
        "\n",
        "#one hot encode Company\n",
        "title = pd.get_dummies(final['Company'], prefix=\"Company\")\n",
        "final = pd.concat([final, title], axis=1)\n",
        "\n",
        "# drop redundant features and save\n",
        "to_drop = [\"Deck\", \"Embarked\", \"Title\", \"Age_binned\", \"Fare_binned\", \"Company\"]\n",
        "final = final.drop(to_drop, axis=1)\n"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "SfBzWV0Ae-i7",
        "outputId": "5d4bcb88-12f0-4e60-aecd-b88a41944cd7",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 261
        }
      },
      "source": [
        "#check final dataframe\n",
        "final.head()"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/html": [
              "<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>PassengerId</th>\n",
              "      <th>Survived</th>\n",
              "      <th>Pclass</th>\n",
              "      <th>Sex</th>\n",
              "      <th>InGroup</th>\n",
              "      <th>WomanChildGroup</th>\n",
              "      <th>Age_S</th>\n",
              "      <th>FareLog_S</th>\n",
              "      <th>Deck_A</th>\n",
              "      <th>Deck_B</th>\n",
              "      <th>Deck_C</th>\n",
              "      <th>Deck_D</th>\n",
              "      <th>Deck_E</th>\n",
              "      <th>Deck_F</th>\n",
              "      <th>Deck_G</th>\n",
              "      <th>Deck_T</th>\n",
              "      <th>Deck_U</th>\n",
              "      <th>Port_C</th>\n",
              "      <th>Port_Q</th>\n",
              "      <th>Port_S</th>\n",
              "      <th>Title_Master</th>\n",
              "      <th>Title_Miss</th>\n",
              "      <th>Title_Mr</th>\n",
              "      <th>Title_Mrs</th>\n",
              "      <th>Title_Other</th>\n",
              "      <th>Age_Child</th>\n",
              "      <th>Age_Teen</th>\n",
              "      <th>Age_ Young Adult</th>\n",
              "      <th>Age_Adult</th>\n",
              "      <th>Age_Senior</th>\n",
              "      <th>Fare_Very Low</th>\n",
              "      <th>Fare_Low</th>\n",
              "      <th>Fare_Average</th>\n",
              "      <th>Fare_Above Average</th>\n",
              "      <th>Fare_High</th>\n",
              "      <th>Fare_Very High</th>\n",
              "      <th>Company_Alone</th>\n",
              "      <th>Company_Large Group</th>\n",
              "      <th>Company_Pair</th>\n",
              "      <th>Company_Small Group</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>0</th>\n",
              "      <td>1</td>\n",
              "      <td>0.0</td>\n",
              "      <td>3</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>-0.528023</td>\n",
              "      <td>-0.879741</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1</th>\n",
              "      <td>2</td>\n",
              "      <td>1.0</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>0.657596</td>\n",
              "      <td>1.361220</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2</th>\n",
              "      <td>3</td>\n",
              "      <td>1.0</td>\n",
              "      <td>3</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>-0.231618</td>\n",
              "      <td>-0.798540</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>3</th>\n",
              "      <td>4</td>\n",
              "      <td>1.0</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>0.435293</td>\n",
              "      <td>1.062038</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>4</th>\n",
              "      <td>5</td>\n",
              "      <td>0.0</td>\n",
              "      <td>3</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0.435293</td>\n",
              "      <td>-0.784179</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>"
            ],
            "text/plain": [
              "   PassengerId  Survived  ...  Company_Pair  Company_Small Group\n",
              "0            1       0.0  ...             0                    0\n",
              "1            2       1.0  ...             1                    0\n",
              "2            3       1.0  ...             0                    0\n",
              "3            4       1.0  ...             1                    0\n",
              "4            5       0.0  ...             0                    0\n",
              "\n",
              "[5 rows x 40 columns]"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 239
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "LU4KlJRsMtSU"
      },
      "source": [
        "##**4 MODELLING**"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "7T7aWJglGInZ"
      },
      "source": [
        "I made a total of 28 submissions to Kaggle. This Random Forest achieved the highest accuracy score on the Public Leaderboard. Overfitting is what most people struggle with in this particular dataset. I kept the maximum depth of the Random Forest relatively low, to reduce the effect of overfitting."
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "bnQ9yIyJMZLV"
      },
      "source": [
        "#split in train and test\n",
        "#final_train = final[final[\"PassengerId\"].isin(train[\"PassengerId\"])]\n",
        "#final_test =  final[final[\"PassengerId\"].isin(test[\"PassengerId\"])]"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "HzCty9A8NEuz"
      },
      "source": [
        "#create train-test-split Full dataframe\n",
        "#X1 = final_train.drop([\"PassengerId\", \"Survived\"], axis=1)\n",
        "#y1 = final_train[\"Survived\"]\n",
        "\n",
        "#X1_train, X1_test, y1_train, y1_test = train_test_split(X1, y1, test_size = 0.2, random_state=42)\n"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "FI0l6Y1unseZ",
        "outputId": "610a4e42-30ad-49a3-ddef-8e6e8ad2f52c",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 294
        }
      },
      "source": [
        "#random_state = 4\n",
        "#model = RandomForestClassifier(random_state=4)\n",
        "#parameters = {\"n_estimators\": [50,100,300,500],\n",
        "              \"max_depth\": [2,4,5]}\n",
        "\n",
        "\n",
        "#kfold = KFold(n_splits= 3, random_state=random_state, shuffle = True)\n",
        "\n",
        "#cv = GridSearchCV(model, param_grid = parameters, cv=kfold, scoring = \"accuracy\")\n",
        "\n",
        "# Fit cv to the training set: \n",
        "#cv.fit(X1_train, y1_train)\n",
        "\n",
        "# Predict on test set\n",
        "#y1_pred = cv.predict(X1_test)\n",
        "#y1_pred_proba = cv.predict_proba(X1_test)[:,1]\n",
        "\n",
        "\n",
        "# Compute and print accuracy\n",
        "#print(\"Accuracy train: {}\".format(cv.score(X1_train, y1_train)))\n",
        "#print(\"Accuracy test: {}\".format(cv.score(X1_test, y1_test)))\n",
        "#print(\"AUC score: {}\".format(roc_auc_score(y1_test, y1_pred_proba)))\n",
        "#print(\"Tuned Model Parameters: {}\".format(cv.best_params_))\n",
        "#print(confusion_matrix(y1_test, y1_pred))\n",
        "#print(classification_report(y1_test, y1_pred))\n",
        "\n"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "Accuracy train: 0.8539325842696629\n",
            "Accuracy test: 0.8268156424581006\n",
            "AUC score: 0.8889961389961388\n",
            "Tuned Model Parameters: {'max_depth': 5, 'n_estimators': 50}\n",
            "[[92 13]\n",
            " [18 56]]\n",
            "              precision    recall  f1-score   support\n",
            "\n",
            "         0.0       0.84      0.88      0.86       105\n",
            "         1.0       0.81      0.76      0.78        74\n",
            "\n",
            "    accuracy                           0.83       179\n",
            "   macro avg       0.82      0.82      0.82       179\n",
            "weighted avg       0.83      0.83      0.83       179\n",
            "\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "GFq8-APWt7h0",
        "outputId": "2d585f7f-876e-459b-9245-3a2d98da896e",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 639
        }
      },
      "source": [
        "\n",
        "#features= X1_train.columns\n",
        "#importances = cv.best_estimator_.feature_importances_\n",
        "#indices = np.argsort(importances)\n",
        "\n",
        "#fig, ax = plt.subplots(figsize=(5,10)) \n",
        "#plt.figure(1)\n",
        "#plt.title('Feature Importances')\n",
        "#plt.barh(range(len(indices)), importances[indices], color='b', align='center')\n",
        "#plt.yticks(range(len(indices)), features[indices])\n",
        "#plt.xlabel('Relative Importance')"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "Text(0.5, 0, 'Relative Importance')"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 463
        },
        {
          "output_type": "display_data",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<Figure size 360x720 with 1 Axes>"
            ]
          },
          "metadata": {
            "tags": [],
            "needs_background": "light"
          }
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "zrTjWtcDw-s-"
      },
      "source": [
        "# make predictions on test data\n",
        "#final_test [\"Survived\"] = cv.predict(final_test.drop([\"PassengerId\", \"Survived\"], axis=1))\n",
        "#final_test [\"Survived\"] = final_test [\"Survived\"].astype(\"int\")\n",
        "\n",
        "\n",
        "# save the submission in the right format\n",
        "#final_test[[\"PassengerId\", \"Survived\"]].to_csv('kaggle_sumbission_F1.csv', index=False)"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "lNkSQEvnIwaj"
      },
      "source": [
        "##**5 CONCLUSION**"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "MakWbSmyI6Rt"
      },
      "source": [
        "\n",
        "**ACCURACY**\n",
        "\n",
        "My testing accuracy was 82.7% and my accuracy on the public leaderboard was 79.2%. In comparison to many of my other models, this model didn't overfit too much.\n",
        "\n",
        "I correctly predicted 92 passengers who didn't survive and 56 passengers who did survive. I had 18 false negatives (I predicted these passengers would survive, but they didn't) and 13 false positives (I predicted that the passenger would live, but they died). \n",
        "\n",
        "Looking at the confusion matrix and classification report, my model seems to be better at predicting who died than who lived. Much more people died than lived, so this isn't too strange. Any feature that could help predict what characteristics helped you survive will probably improve the model.\n",
        "\n",
        "\n",
        "**FEATURE IMPORTANCE**\n",
        "\n",
        "The feature importance graph indicates that the following features contributed most to the decision making process. \n",
        "\n",
        "\n",
        "1.   Gender\n",
        "2.   Mr Title (adult male)\n",
        "3.   Mrs Title (female passenger of any age)\n",
        "4.   Class\n",
        "5.   Fare (log transformed)\n",
        "6.   Title Miss (unmarried woman, but in practice females of any age)\n",
        "7.   Age (scaled)\n",
        "8.   Travelling in Large groups (5 or more persons)\n",
        "9.   Travelling in Woman Child group (group with >1 woman or children)\n",
        "10.  Having an unknown deck (often third class Passenger)\n",
        "\n",
        "\n",
        "**FUTURE IMPROVEMENTS**\n",
        "\n",
        "It is clear that gender, age, class and travel company are important factors in the decision-making process. Age is a feature with a lot of missing values, so the better you can impute these values, the better the model is. Looking at Mice or KNN imputing could be an interesting idea. A lot of people have mislabeled titles, fixing these might improve the model further. Correctly identifying who is travelling with whom is also a challenge that could improve the peformance of the model. And then of course there is hyperparameter tuning (although in my case this often resulted in a lower score on the public leaderboard) and stacking different models together.\n",
        "\n",
        "\n",
        "**CONCLUDING REMARK**\n",
        "\n",
        "Although the Titanic sank over 100 years ago, it still captivates many people, including many data scientists. New features and models still appear very regularly. Pretty fascinating stuff! "
      ]
    }
  ]
}