q3.ipynb

{
  "nbformat": 4,
  "nbformat_minor": 0,
  "metadata": {
    "kernelspec": {
      "display_name": "Python 3",
      "language": "python",
      "name": "python3"
    },
    "language_info": {
      "codemirror_mode": {
        "name": "ipython",
        "version": 3
      },
      "file_extension": ".py",
      "mimetype": "text/x-python",
      "name": "python",
      "nbconvert_exporter": "python",
      "pygments_lexer": "ipython3",
      "version": "3.7.6"
    },
    "colab": {
      "name": "sol3.ipynb",
      "provenance": [],
      "collapsed_sections": []
    }
  },
  "cells": [
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "1OzpIxVjs0EJ"
      },
      "source": [
        "\n",
        "1. You are provided the titanic dataset. Load the dataset and perform splitting into training and test sets with 70:30 ratio randomly using test train split.\n",
        "2. Use the Logistic regression created from scratch (from the prev question) in this question as well.\n",
        "3. Data cleaning plays a major role in this question. Report all the methods used by you in the ipynb.\n",
        "\n",
        "--> \n",
        "\n",
        "i. Check for missing values\n",
        "\n",
        "ii. Drop Columns & Handle missing values\n",
        "\n",
        "iii. Create dummies for categorical features\n",
        "\n",
        "you are free to perform other data cleaning to improve your results.\n",
        "\n",
        "\n",
        "\n",
        "4. Report accuracy score, Confusion matrix, heat map, classifiaction report and any other metrics you feel useful."
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "xhEKd9yqs0ES"
      },
      "source": [
        "dataset link : \n",
        "https://iiitaphyd-my.sharepoint.com/:f:/g/personal/apurva_jadhav_students_iiit_ac_in/Eictt5_qmoxNqezgQQiMWeIBph4sxlfA6jWAJNPnV2SF9Q?e=mQmYN0 \n",
        "\n",
        "(titanic.csv)\n"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "PRcYL9_2s0ET",
        "outputId": "a82b6f6c-55ec-4477-d088-8fe9b1c2d9f9"
      },
      "source": [
        "from google.colab import drive\n",
        "drive.mount('/content/drive')\n",
        "%cd /content/drive/MyDrive/SMAI/Assignment2/q_datasets\n",
        "!pwd\n",
        "!ls"
      ],
      "execution_count": 1,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "Mounted at /content/drive\n",
            "/content/drive/MyDrive/SMAI/Assignment2/q_datasets\n",
            "/content/drive/MyDrive/SMAI/Assignment2/q_datasets\n",
            "'archive (1).zip'   fashion_mnist   titanic.csv   X.npy   Y.npy\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "BzVzNzGItEx2"
      },
      "source": [
        "import pandas as pd\n",
        "import numpy as np\n",
        "from sklearn.model_selection import train_test_split\n",
        "from sklearn.preprocessing import StandardScaler\n",
        "from sklearn.preprocessing import Normalizer\n",
        "import matplotlib.pyplot as plt\n",
        "from sklearn.metrics import r2_score, mean_absolute_error, mean_squared_error\n",
        "from sklearn.linear_model import LogisticRegression\n",
        "from sklearn import preprocessing\n",
        "from sklearn.metrics import confusion_matrix\n",
        "import seaborn as sn\n",
        "from sklearn.preprocessing import OneHotEncoder"
      ],
      "execution_count": 2,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 841
        },
        "id": "4_SjUl59uMFl",
        "outputId": "8bc0b21a-edfc-4088-a2a4-f5b3e3b67cda"
      },
      "source": [
        "df_data_org = pd.read_csv('titanic.csv', na_values=np.NaN)\n",
        "print(df_data_org.shape)\n",
        "display(df_data_org.head())\n",
        "display(df_data_org.info())\n",
        "display(df_data_org.describe())"
      ],
      "execution_count": 28,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "(891, 12)\n"
          ],
          "name": "stdout"
        },
        {
          "output_type": "display_data",
          "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</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</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</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</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</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       3  ...   7.2500   NaN         S\n",
              "1            2         1       1  ...  71.2833   C85         C\n",
              "2            3         1       3  ...   7.9250   NaN         S\n",
              "3            4         1       1  ...  53.1000  C123         S\n",
              "4            5         0       3  ...   8.0500   NaN         S\n",
              "\n",
              "[5 rows x 12 columns]"
            ]
          },
          "metadata": {
            "tags": []
          }
        },
        {
          "output_type": "stream",
          "text": [
            "<class 'pandas.core.frame.DataFrame'>\n",
            "RangeIndex: 891 entries, 0 to 890\n",
            "Data columns (total 12 columns):\n",
            " #   Column       Non-Null Count  Dtype  \n",
            "---  ------       --------------  -----  \n",
            " 0   PassengerId  891 non-null    int64  \n",
            " 1   Survived     891 non-null    int64  \n",
            " 2   Pclass       891 non-null    int64  \n",
            " 3   Name         891 non-null    object \n",
            " 4   Sex          891 non-null    object \n",
            " 5   Age          714 non-null    float64\n",
            " 6   SibSp        891 non-null    int64  \n",
            " 7   Parch        891 non-null    int64  \n",
            " 8   Ticket       891 non-null    object \n",
            " 9   Fare         891 non-null    float64\n",
            " 10  Cabin        204 non-null    object \n",
            " 11  Embarked     889 non-null    object \n",
            "dtypes: float64(2), int64(5), object(5)\n",
            "memory usage: 83.7+ KB\n"
          ],
          "name": "stdout"
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "None"
            ]
          },
          "metadata": {
            "tags": []
          }
        },
        {
          "output_type": "display_data",
          "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>Age</th>\n",
              "      <th>SibSp</th>\n",
              "      <th>Parch</th>\n",
              "      <th>Fare</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>count</th>\n",
              "      <td>891.000000</td>\n",
              "      <td>891.000000</td>\n",
              "      <td>891.000000</td>\n",
              "      <td>714.000000</td>\n",
              "      <td>891.000000</td>\n",
              "      <td>891.000000</td>\n",
              "      <td>891.000000</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>mean</th>\n",
              "      <td>446.000000</td>\n",
              "      <td>0.383838</td>\n",
              "      <td>2.308642</td>\n",
              "      <td>29.699118</td>\n",
              "      <td>0.523008</td>\n",
              "      <td>0.381594</td>\n",
              "      <td>32.204208</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>std</th>\n",
              "      <td>257.353842</td>\n",
              "      <td>0.486592</td>\n",
              "      <td>0.836071</td>\n",
              "      <td>14.526497</td>\n",
              "      <td>1.102743</td>\n",
              "      <td>0.806057</td>\n",
              "      <td>49.693429</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>min</th>\n",
              "      <td>1.000000</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>1.000000</td>\n",
              "      <td>0.420000</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.000000</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>25%</th>\n",
              "      <td>223.500000</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>2.000000</td>\n",
              "      <td>20.125000</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>7.910400</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>50%</th>\n",
              "      <td>446.000000</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>3.000000</td>\n",
              "      <td>28.000000</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>14.454200</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>75%</th>\n",
              "      <td>668.500000</td>\n",
              "      <td>1.000000</td>\n",
              "      <td>3.000000</td>\n",
              "      <td>38.000000</td>\n",
              "      <td>1.000000</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>31.000000</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>max</th>\n",
              "      <td>891.000000</td>\n",
              "      <td>1.000000</td>\n",
              "      <td>3.000000</td>\n",
              "      <td>80.000000</td>\n",
              "      <td>8.000000</td>\n",
              "      <td>6.000000</td>\n",
              "      <td>512.329200</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>"
            ],
            "text/plain": [
              "       PassengerId    Survived      Pclass  ...       SibSp       Parch        Fare\n",
              "count   891.000000  891.000000  891.000000  ...  891.000000  891.000000  891.000000\n",
              "mean    446.000000    0.383838    2.308642  ...    0.523008    0.381594   32.204208\n",
              "std     257.353842    0.486592    0.836071  ...    1.102743    0.806057   49.693429\n",
              "min       1.000000    0.000000    1.000000  ...    0.000000    0.000000    0.000000\n",
              "25%     223.500000    0.000000    2.000000  ...    0.000000    0.000000    7.910400\n",
              "50%     446.000000    0.000000    3.000000  ...    0.000000    0.000000   14.454200\n",
              "75%     668.500000    1.000000    3.000000  ...    1.000000    0.000000   31.000000\n",
              "max     891.000000    1.000000    3.000000  ...    8.000000    6.000000  512.329200\n",
              "\n",
              "[8 rows x 7 columns]"
            ]
          },
          "metadata": {
            "tags": []
          }
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "-YrpTV44uQ8G",
        "outputId": "97ba2cfe-018c-4a4f-c5ac-c7515aa3ad56"
      },
      "source": [
        "# checking missing valued columns with counts\n",
        "missing_val_count_by_column = (df_data_org.isnull().sum())\n",
        "print(missing_val_count_by_column[missing_val_count_by_column > 0])"
      ],
      "execution_count": 29,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "Age         177\n",
            "Cabin       687\n",
            "Embarked      2\n",
            "dtype: int64\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "r4Z1u69iudN4",
        "outputId": "19bbfe9f-d5ad-4a0a-a8ec-7eaf4b26ff95"
      },
      "source": [
        "### Embarked has only 2 null rows, so remove those rows\n",
        "df_data_org = df_data_org[df_data_org['Embarked'].notna()]\n",
        "print(df_data_org.shape)\n",
        "missing_val_count_by_column = (df_data_org.isnull().sum())\n",
        "print(missing_val_count_by_column[missing_val_count_by_column > 0])"
      ],
      "execution_count": 30,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "(889, 12)\n",
            "Age      177\n",
            "Cabin    687\n",
            "dtype: int64\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 408
        },
        "id": "fdlZ-aazuiZo",
        "outputId": "e694ead8-22b2-4052-f652-0519416b75e5"
      },
      "source": [
        "### Replace missing age with median age (to keep the distribution same) of each (pclass, sex) group\n",
        "df_data_org['Age'] = df_data_org.groupby(['Sex','Pclass'])['Age'].apply(lambda a: a.fillna(a.median()))\n",
        "print(df_data_org.shape)\n",
        "missing_val_count_by_column = (df_data_org.isnull().sum())\n",
        "print(missing_val_count_by_column[missing_val_count_by_column > 0])\n",
        "display(df_data_org.info())"
      ],
      "execution_count": 31,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "(889, 12)\n",
            "Cabin    687\n",
            "dtype: int64\n",
            "<class 'pandas.core.frame.DataFrame'>\n",
            "Int64Index: 889 entries, 0 to 890\n",
            "Data columns (total 12 columns):\n",
            " #   Column       Non-Null Count  Dtype  \n",
            "---  ------       --------------  -----  \n",
            " 0   PassengerId  889 non-null    int64  \n",
            " 1   Survived     889 non-null    int64  \n",
            " 2   Pclass       889 non-null    int64  \n",
            " 3   Name         889 non-null    object \n",
            " 4   Sex          889 non-null    object \n",
            " 5   Age          889 non-null    float64\n",
            " 6   SibSp        889 non-null    int64  \n",
            " 7   Parch        889 non-null    int64  \n",
            " 8   Ticket       889 non-null    object \n",
            " 9   Fare         889 non-null    float64\n",
            " 10  Cabin        202 non-null    object \n",
            " 11  Embarked     889 non-null    object \n",
            "dtypes: float64(2), int64(5), object(5)\n",
            "memory usage: 130.3+ KB\n"
          ],
          "name": "stdout"
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "None"
            ]
          },
          "metadata": {
            "tags": []
          }
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "6N8WppH8vN1M"
      },
      "source": [
        "Though cabin has less non-null rows, the floor on which a person stayed matters as it tells if he could reach the top-floor quickly or no, So extacted only the floow (1st char) of the cabin number"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 391
        },
        "id": "GAUfHpvsutZr",
        "outputId": "d77b0ca1-90bf-4501-a74a-62ff1a9f73cb"
      },
      "source": [
        "df_data_org['Cabin_floor'] = df_data_org['Cabin'].apply(lambda c: c[0] if pd.notnull(c) else 'M')\n",
        "df_data_org = df_data_org.drop(['Cabin'], axis=1)\n",
        "print(df_data_org.shape)\n",
        "missing_val_count_by_column = (df_data_org.isnull().sum())\n",
        "print(missing_val_count_by_column[missing_val_count_by_column > 0])\n",
        "display(df_data_org.info())"
      ],
      "execution_count": 32,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "(889, 12)\n",
            "Series([], dtype: int64)\n",
            "<class 'pandas.core.frame.DataFrame'>\n",
            "Int64Index: 889 entries, 0 to 890\n",
            "Data columns (total 12 columns):\n",
            " #   Column       Non-Null Count  Dtype  \n",
            "---  ------       --------------  -----  \n",
            " 0   PassengerId  889 non-null    int64  \n",
            " 1   Survived     889 non-null    int64  \n",
            " 2   Pclass       889 non-null    int64  \n",
            " 3   Name         889 non-null    object \n",
            " 4   Sex          889 non-null    object \n",
            " 5   Age          889 non-null    float64\n",
            " 6   SibSp        889 non-null    int64  \n",
            " 7   Parch        889 non-null    int64  \n",
            " 8   Ticket       889 non-null    object \n",
            " 9   Fare         889 non-null    float64\n",
            " 10  Embarked     889 non-null    object \n",
            " 11  Cabin_floor  889 non-null    object \n",
            "dtypes: float64(2), int64(5), object(5)\n",
            "memory usage: 130.3+ KB\n"
          ],
          "name": "stdout"
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "None"
            ]
          },
          "metadata": {
            "tags": []
          }
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "pVYcz8pYyz9N"
      },
      "source": [
        "Dropping columns passenger-id, name and ticket as they are unique for every person and do dot contain any specific information"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 527
        },
        "id": "oNy6jD47yyTw",
        "outputId": "86acad83-9e33-4693-ff6b-c1578adee38c"
      },
      "source": [
        "df_data_org = df_data_org.drop(['PassengerId', 'Name', 'Ticket'], axis=1)\n",
        "print(df_data_org.shape)\n",
        "missing_val_count_by_column = (df_data_org.isnull().sum())\n",
        "print(missing_val_count_by_column[missing_val_count_by_column > 0])\n",
        "display(df_data_org.info())\n",
        "display(df_data_org.head())"
      ],
      "execution_count": 33,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "(889, 9)\n",
            "Series([], dtype: int64)\n",
            "<class 'pandas.core.frame.DataFrame'>\n",
            "Int64Index: 889 entries, 0 to 890\n",
            "Data columns (total 9 columns):\n",
            " #   Column       Non-Null Count  Dtype  \n",
            "---  ------       --------------  -----  \n",
            " 0   Survived     889 non-null    int64  \n",
            " 1   Pclass       889 non-null    int64  \n",
            " 2   Sex          889 non-null    object \n",
            " 3   Age          889 non-null    float64\n",
            " 4   SibSp        889 non-null    int64  \n",
            " 5   Parch        889 non-null    int64  \n",
            " 6   Fare         889 non-null    float64\n",
            " 7   Embarked     889 non-null    object \n",
            " 8   Cabin_floor  889 non-null    object \n",
            "dtypes: float64(2), int64(4), object(3)\n",
            "memory usage: 109.5+ KB\n"
          ],
          "name": "stdout"
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "None"
            ]
          },
          "metadata": {
            "tags": []
          }
        },
        {
          "output_type": "display_data",
          "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>Survived</th>\n",
              "      <th>Pclass</th>\n",
              "      <th>Sex</th>\n",
              "      <th>Age</th>\n",
              "      <th>SibSp</th>\n",
              "      <th>Parch</th>\n",
              "      <th>Fare</th>\n",
              "      <th>Embarked</th>\n",
              "      <th>Cabin_floor</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>0</th>\n",
              "      <td>0</td>\n",
              "      <td>3</td>\n",
              "      <td>male</td>\n",
              "      <td>22.0</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>7.2500</td>\n",
              "      <td>S</td>\n",
              "      <td>M</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1</th>\n",
              "      <td>1</td>\n",
              "      <td>1</td>\n",
              "      <td>female</td>\n",
              "      <td>38.0</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>71.2833</td>\n",
              "      <td>C</td>\n",
              "      <td>C</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2</th>\n",
              "      <td>1</td>\n",
              "      <td>3</td>\n",
              "      <td>female</td>\n",
              "      <td>26.0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>7.9250</td>\n",
              "      <td>S</td>\n",
              "      <td>M</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>3</th>\n",
              "      <td>1</td>\n",
              "      <td>1</td>\n",
              "      <td>female</td>\n",
              "      <td>35.0</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>53.1000</td>\n",
              "      <td>S</td>\n",
              "      <td>C</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>4</th>\n",
              "      <td>0</td>\n",
              "      <td>3</td>\n",
              "      <td>male</td>\n",
              "      <td>35.0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>8.0500</td>\n",
              "      <td>S</td>\n",
              "      <td>M</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>"
            ],
            "text/plain": [
              "   Survived  Pclass     Sex   Age  SibSp  Parch     Fare Embarked Cabin_floor\n",
              "0         0       3    male  22.0      1      0   7.2500        S           M\n",
              "1         1       1  female  38.0      1      0  71.2833        C           C\n",
              "2         1       3  female  26.0      0      0   7.9250        S           M\n",
              "3         1       1  female  35.0      1      0  53.1000        S           C\n",
              "4         0       3    male  35.0      0      0   8.0500        S           M"
            ]
          },
          "metadata": {
            "tags": []
          }
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "CN_1R1S2ztOu"
      },
      "source": [
        "Converting to one-hot encoding for classes: Sex, Cabin_floor and Embarked. Ordinal encoding is not used for Embarked as the ship sank before anyone could reach the destination so there is no order. For cabin_floor, many values were missing so we cannot assign proper order for missing category so we keep it one-hot encoded"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 700
        },
        "id": "b9HDBRr4zCON",
        "outputId": "422ba3eb-6ed7-4a58-8731-37652243e7be"
      },
      "source": [
        "cat_to_num = {\"Sex\":{'female':0, 'male':1}}\n",
        "df_data_org = df_data_org.replace(cat_to_num)\n",
        "\n",
        "enc_feats = []\n",
        "\n",
        "for feat in ['Cabin_floor', 'Embarked']:\n",
        "    onehot_enc = OneHotEncoder().fit_transform(df_data_org[feat].values.reshape(-1,1)).toarray()\n",
        "    n = df_data_org[feat].nunique()\n",
        "    colnames = [feat+'_'+str(i) for i in range(n)]\n",
        "    df_onehotenc = pd.DataFrame(onehot_enc, columns=colnames, index=df_data_org.index)\n",
        "    enc_feats.append(df_onehotenc)\n",
        "    df_data_org.drop([feat], axis=1, inplace=True)\n",
        "\n",
        "df_data_org = pd.concat([df_data_org, *enc_feats], axis=1)  \n",
        "  \n",
        "\n",
        "print(df_data_org.shape)\n",
        "display(df_data_org.head())\n",
        "display(df_data_org.info())"
      ],
      "execution_count": 34,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "(889, 19)\n"
          ],
          "name": "stdout"
        },
        {
          "output_type": "display_data",
          "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>Survived</th>\n",
              "      <th>Pclass</th>\n",
              "      <th>Sex</th>\n",
              "      <th>Age</th>\n",
              "      <th>SibSp</th>\n",
              "      <th>Parch</th>\n",
              "      <th>Fare</th>\n",
              "      <th>Cabin_floor_0</th>\n",
              "      <th>Cabin_floor_1</th>\n",
              "      <th>Cabin_floor_2</th>\n",
              "      <th>Cabin_floor_3</th>\n",
              "      <th>Cabin_floor_4</th>\n",
              "      <th>Cabin_floor_5</th>\n",
              "      <th>Cabin_floor_6</th>\n",
              "      <th>Cabin_floor_7</th>\n",
              "      <th>Cabin_floor_8</th>\n",
              "      <th>Embarked_0</th>\n",
              "      <th>Embarked_1</th>\n",
              "      <th>Embarked_2</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>0</th>\n",
              "      <td>0</td>\n",
              "      <td>3</td>\n",
              "      <td>1</td>\n",
              "      <td>22.0</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>7.2500</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>1.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>1.0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1</th>\n",
              "      <td>1</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>38.0</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>71.2833</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>1.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>1.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2</th>\n",
              "      <td>1</td>\n",
              "      <td>3</td>\n",
              "      <td>0</td>\n",
              "      <td>26.0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>7.9250</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>1.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>1.0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>3</th>\n",
              "      <td>1</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>35.0</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>53.1000</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>1.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>1.0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>4</th>\n",
              "      <td>0</td>\n",
              "      <td>3</td>\n",
              "      <td>1</td>\n",
              "      <td>35.0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>8.0500</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>1.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>1.0</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>"
            ],
            "text/plain": [
              "   Survived  Pclass  Sex  ...  Embarked_0  Embarked_1  Embarked_2\n",
              "0         0       3    1  ...         0.0         0.0         1.0\n",
              "1         1       1    0  ...         1.0         0.0         0.0\n",
              "2         1       3    0  ...         0.0         0.0         1.0\n",
              "3         1       1    0  ...         0.0         0.0         1.0\n",
              "4         0       3    1  ...         0.0         0.0         1.0\n",
              "\n",
              "[5 rows x 19 columns]"
            ]
          },
          "metadata": {
            "tags": []
          }
        },
        {
          "output_type": "stream",
          "text": [
            "<class 'pandas.core.frame.DataFrame'>\n",
            "Int64Index: 889 entries, 0 to 890\n",
            "Data columns (total 19 columns):\n",
            " #   Column         Non-Null Count  Dtype  \n",
            "---  ------         --------------  -----  \n",
            " 0   Survived       889 non-null    int64  \n",
            " 1   Pclass         889 non-null    int64  \n",
            " 2   Sex            889 non-null    int64  \n",
            " 3   Age            889 non-null    float64\n",
            " 4   SibSp          889 non-null    int64  \n",
            " 5   Parch          889 non-null    int64  \n",
            " 6   Fare           889 non-null    float64\n",
            " 7   Cabin_floor_0  889 non-null    float64\n",
            " 8   Cabin_floor_1  889 non-null    float64\n",
            " 9   Cabin_floor_2  889 non-null    float64\n",
            " 10  Cabin_floor_3  889 non-null    float64\n",
            " 11  Cabin_floor_4  889 non-null    float64\n",
            " 12  Cabin_floor_5  889 non-null    float64\n",
            " 13  Cabin_floor_6  889 non-null    float64\n",
            " 14  Cabin_floor_7  889 non-null    float64\n",
            " 15  Cabin_floor_8  889 non-null    float64\n",
            " 16  Embarked_0     889 non-null    float64\n",
            " 17  Embarked_1     889 non-null    float64\n",
            " 18  Embarked_2     889 non-null    float64\n",
            "dtypes: float64(14), int64(5)\n",
            "memory usage: 178.9 KB\n"
          ],
          "name": "stdout"
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "None"
            ]
          },
          "metadata": {
            "tags": []
          }
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "N_Tf3aAyz17R"
      },
      "source": [
        "Removing the last columns of Cabin_floor and Embarked, as all zeros implies these categories"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 459
        },
        "id": "j7pOSzs1zWhe",
        "outputId": "d9779500-6abf-4b65-8441-d54405cca186"
      },
      "source": [
        "df_data_org = df_data_org.drop(['Cabin_floor_8', 'Embarked_2'], axis=1)\n",
        "print(df_data_org.shape)\n",
        "display(df_data_org.info())"
      ],
      "execution_count": 35,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "(889, 17)\n",
            "<class 'pandas.core.frame.DataFrame'>\n",
            "Int64Index: 889 entries, 0 to 890\n",
            "Data columns (total 17 columns):\n",
            " #   Column         Non-Null Count  Dtype  \n",
            "---  ------         --------------  -----  \n",
            " 0   Survived       889 non-null    int64  \n",
            " 1   Pclass         889 non-null    int64  \n",
            " 2   Sex            889 non-null    int64  \n",
            " 3   Age            889 non-null    float64\n",
            " 4   SibSp          889 non-null    int64  \n",
            " 5   Parch          889 non-null    int64  \n",
            " 6   Fare           889 non-null    float64\n",
            " 7   Cabin_floor_0  889 non-null    float64\n",
            " 8   Cabin_floor_1  889 non-null    float64\n",
            " 9   Cabin_floor_2  889 non-null    float64\n",
            " 10  Cabin_floor_3  889 non-null    float64\n",
            " 11  Cabin_floor_4  889 non-null    float64\n",
            " 12  Cabin_floor_5  889 non-null    float64\n",
            " 13  Cabin_floor_6  889 non-null    float64\n",
            " 14  Cabin_floor_7  889 non-null    float64\n",
            " 15  Embarked_0     889 non-null    float64\n",
            " 16  Embarked_1     889 non-null    float64\n",
            "dtypes: float64(12), int64(5)\n",
            "memory usage: 165.0 KB\n"
          ],
          "name": "stdout"
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "None"
            ]
          },
          "metadata": {
            "tags": []
          }
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "dDIBG-fz3pih"
      },
      "source": [
        "No 2 columns have very high correlation, so we take all"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 344
        },
        "id": "ZL7kMXI927Uj",
        "outputId": "34136a16-7528-431a-c675-1cf1287e4d9d"
      },
      "source": [
        "corr = df_data_org.corr()\n",
        "sn.heatmap(corr)"
      ],
      "execution_count": 38,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "<matplotlib.axes._subplots.AxesSubplot at 0x7f00eb1c3f50>"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 38
        },
        {
          "output_type": "display_data",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<Figure size 432x288 with 2 Axes>"
            ]
          },
          "metadata": {
            "tags": [],
            "needs_background": "light"
          }
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "LpYoRDPt2mVA"
      },
      "source": [
        "As columns have different scales, we standardize the data to keep the relative distances within features same"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "889H0Kc-0N36",
        "outputId": "cc04ba84-8749-4704-e711-ecfdb33d3097"
      },
      "source": [
        "X = np.array(df_data_org.drop('Survived', axis=1))\n",
        "y = np.array(df_data_org.Survived)\n",
        "print(len(X))\n",
        "print(len(y))\n",
        "\n",
        "\n",
        "X_train_org, X_test_org, y_train, y_test = train_test_split(X, y, test_size=0.30,random_state=0)\n",
        "print(X_train_org.shape, y_train.shape)\n",
        "print(X_test_org.shape, y_test.shape)\n",
        "\n",
        "'''Standardize X'''\n",
        "scaler = StandardScaler()\n",
        "X_train_std = scaler.fit_transform(X_train_org)\n",
        "X_test_std = scaler.transform(X_test_org)\n"
      ],
      "execution_count": 39,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "889\n",
            "889\n",
            "(622, 16) (622,)\n",
            "(267, 16) (267,)\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "mrVxOo5S3GiU"
      },
      "source": [
        "class MyLogRegressor(object):\n",
        "    def __init__(self, lr=0.2, itrs=100):\n",
        "        self.lr = lr \n",
        "        self.itrs = itrs\n",
        "        self.W = None\n",
        "        self.cost = None\n",
        "        self.method = None\n",
        "    \n",
        "    def fit(self, X, Y):\n",
        "        X = np.insert(X, 0, 1, axis=1)\n",
        "        n = len(Y)\n",
        "        W = np.zeros(X.shape[1])\n",
        "        cost = []\n",
        "        for i in range(self.itrs):\n",
        "            Z = X.dot(W)\n",
        "            H = self.sigmoid(Z)\n",
        "            W = self.gradient_desc(X,H,W,Y,n)\n",
        "            cost.append(self.cal_cost(H,W,Y)) \n",
        "        self.W = W\n",
        "        self.cost = cost\n",
        "        return self\n",
        "\n",
        "    def gradient_desc(self,X,H,W,Y,n):\n",
        "        gradient = np.dot(X.T, (H - Y)) / n\n",
        "        W = W - self.lr * gradient\n",
        "        return W \n",
        "\n",
        "    def sigmoid(self, Z):\n",
        "        return 1 / (1 + np.exp(-Z))\n",
        "\n",
        "    def cal_cost(self,H,W,Y):\n",
        "        n = len(Y)\n",
        "        cost = (np.sum(-Y.T.dot(np.log(H)) - (1 - Y).T.dot(np.log(1 - H)))) / n\n",
        "        return cost\n",
        "\n",
        "    def predict(self, X):\n",
        "        X = np.insert(X, 0, 1, axis=1)\n",
        "        Y_pred = [round(self.sigmoid(i.dot(self.W))) for i in X ]\n",
        "        return Y_pred\n",
        "\n",
        "    def cal_score(self,X, Y):\n",
        "        score = sum(self.predict(X) == Y) / len(Y)\n",
        "        return score\n",
        "    \n",
        "    def plot_cost(self): \n",
        "        plt.plot(range(self.itrs), self.cost)\n",
        "        plt.xlabel('Iterations')\n",
        "        plt.ylabel('Cost')\n",
        "        plt.title('Convergence Graph of Cost Function')\n",
        "        # plt.legend()\n",
        "        plt.show()\n"
      ],
      "execution_count": 42,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 577
        },
        "id": "sPgcv64W4lTG",
        "outputId": "0408fd4d-887e-4e48-b64b-243d0d4727fa"
      },
      "source": [
        "itrs = 500\n",
        "logreg = MyLogRegressor(lr=0.2, itrs=itrs).fit(X_train_std, y_train)\n",
        "logreg.plot_cost()\n",
        "\n",
        "myscore1 = logreg.cal_score(X_train_std,y_train)\n",
        "y_pred1 = logreg.predict(X_train_std)\n",
        "print(\"Training accuracy:\",myscore1)\n",
        "conf_mat1 = confusion_matrix(y_train, y_pred1)\n",
        "df_cm1 = pd.DataFrame(conf_mat1, index = [i for i in range(2)], columns = ['female', 'male'])\n",
        "plt.figure()\n",
        "sn.heatmap(df_cm1, annot=True, cmap='Blues')"
      ],
      "execution_count": 45,
      "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"
          }
        },
        {
          "output_type": "stream",
          "text": [
            "Training accuracy: 0.8279742765273312\n"
          ],
          "name": "stdout"
        },
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "<matplotlib.axes._subplots.AxesSubplot at 0x7f00e2386fd0>"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 45
        },
        {
          "output_type": "display_data",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<Figure size 432x288 with 2 Axes>"
            ]
          },
          "metadata": {
            "tags": [],
            "needs_background": "light"
          }
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 299
        },
        "id": "yPe7R9Ab43_u",
        "outputId": "be0cd997-1e01-4d83-a0bc-4e8dffeb8a95"
      },
      "source": [
        "myscore2 = logreg.cal_score(X_test_std,y_test)\n",
        "y_pred2 = logreg.predict(X_test_std)\n",
        "print(\"Testing accuracy:\",myscore2)\n",
        "conf_mat2 = confusion_matrix(y_test, y_pred2)\n",
        "df_cm2 = pd.DataFrame(conf_mat2, index = [i for i in range(2)], columns = ['female', 'male'])\n",
        "plt.figure()\n",
        "sn.heatmap(df_cm2, annot=True, cmap='Blues')"
      ],
      "execution_count": 46,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "Testing accuracy: 0.7940074906367042\n"
          ],
          "name": "stdout"
        },
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "<matplotlib.axes._subplots.AxesSubplot at 0x7f00e2086f50>"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 46
        },
        {
          "output_type": "display_data",
          "data": {
            "image/png": "iVBORw0KGgoAAAANSUhEUgAAAWAAAAD4CAYAAADSIzzWAAAABHNCSVQICAgIfAhkiAAAAAlwSFlzAAALEgAACxIB0t1+/AAAADh0RVh0U29mdHdhcmUAbWF0cGxvdGxpYiB2ZXJzaW9uMy4yLjIsIGh0dHA6Ly9tYXRwbG90bGliLm9yZy+WH4yJAAAVYklEQVR4nO3deZgU1b3G8e87MyIMoCDIImgERVEIiCKK28WA+wJxiysgJBOjgksMajSShGhiNHGJ13iJqBiMaDDGJUblEoxXVNxQUEFFUAOCgKwBRAbO/WNanMAw09MzPWeqeD/PUw/dVdWnTms/Lz9OnapSCAEzM6t7BbE7YGa2rXIAm5lF4gA2M4vEAWxmFokD2MwskqJ8H6BRj4s9zcK2sOzVO2J3weqhhkWopm1UJ3PWTrujxserCVfAZmaR5L0CNjOrU0pOXekANrN0KSiM3YOsOYDNLF0UdVi3WhzAZpYuHoIwM4vEFbCZWSSugM3MInEFbGYWiWdBmJlF4iEIM7NIPARhZhaJK2Azs0gcwGZmkRT6JJyZWRweAzYzi8RDEGZmkbgCNjOLxBWwmVkkroDNzCLxpchmZpF4CMLMLBIPQZiZReIK2MwskgQFcHJ6amaWjYLC7JcqSLpH0iJJb5dbd5OkWZKmS3pUUrNy266WNFvSe5KOqbKrOX9JM7P6SMp+qdp9wLGbrZsIdA0hdAPeB64uO6z2Bc4EumQ+c6ekSlPeAWxm6aKC7JcqhBCeB5Zutu7ZEEJp5u3LQPvM6/7A+BDCuhDCXGA20Kuy9h3AZpYu1aiAJZVIeq3cUlLNow0B/p553Q74V7lt8zLrtson4cwsVVSNaWghhNHA6ByPcw1QCjyQy+fBAWxmKVOdAK7BMQYDJwJ9Qwghs3o+sGu53dpn1m2VhyDMLFVUoKyXnNqXjgVGACeHENaU2/Q4cKak7SV1ADoBr1TWlitgM0uV2qyAJT0I9AFaSpoHjKRs1sP2wMTMsV4OIVwQQnhH0sPAu5QNTVwUQthQWfsOYDNLldoM4BDCWRWsHlPJ/tcD12fbvgPYzFKlLsaAa4sD2MzSJTn56wA2s3RxBWxmFklBQXImdzmAzSxVXAGbmcWSnPx1AJtZurgCNjOLxAFsZhZJrpcYx+AANrNUcQVsZhaJA9jMLBIHsJlZJA5gM7NYkpO/DmAzSxdfimxmFomHIMzMYklO/jqAK3PXyHM47oiuLF66ip6n37DF9hP7fJPrfnAiG0OgdMNGRtw0gRffnFOjYzbfoZg/3jiEb+yyEx9/upRzR4xh+aq1nHlcTy4ffBSS+PeaLxh+w0PMeL/S5/1ZPbRwwQKuuXoESz//HCROO/0MzjlvEHfcfivPTZ5EgQpo3qIFo67/Ja1atY7d3URKUgWsrx/omR+Nelyc3wPk0aH778HqNeu4e9TACgO4caMGrF77JQBdO+3CuBuHsN8pv8iq7cMP6MR5Jx9Eychx/7H++kv6s2zlGm6+dyJXnH8UzZoWc+3tj3Fw9w7MmrOQ5avWcvSh+3Lt94/niIE31/xLRrLs1TtidyGKxYsXsWTxYvbZtwurV/+bM08/lVtv/29at2lDkyZNAHhg3P3M+XA2Pxn588i9rXsNi2pev35j+BNZZ87Ht58UNa2rHK2W1FnSlZJuzyxXStqnLjoX25Q3PmTpijVb3f5V+AI0brQ95f8uu2xgX14Y9yNeeehqrr3g+KyPeWKfbox7YioA456YyklHdgPg5bfmsnzVWgBemT6Xdq2bVeerWD2x886t2GffLgA0btyEjh07smjRZ5vCF+CLtWsTVcXVN5KyXmKrdAhC0pXAWcB4vn68cnvgQUnjQwi/ynP/6r2Tj+zGz4edzM47NeWU4XcB0PfgzuyxWysOO/cmJDHh1u9z6P57MOWND6tsr1WLpixcshKAhUtW0qpF0y32GTzgEJ6Z8m7tfhGrc/Pnz2PWzJl8s1t3AH532y088fhfadKkKXffe3/k3iVXmu4FMRToEkJYX36lpN8C7wAVBrCkEqAEoKh9H4padqmFrtZPj0+ezuOTp3Po/ntw3YUncMIFd9Cv9z70692Zl8dfBUCTRtuz526tmPLGhzx//xU0aFBEk0bb03zH4k37XHvbY/zvSzO3aH/zEaIjenZi0IDe9B1yS96/m+XPmtWr+eGlw/nRVT/eVP0Ou+Qyhl1yGWP+8D+M/9M4Lrx4eOReJlN9qGyzVVUAbwR2AT7ebH3bzLYKhRBGA6Mh2WPA1THljQ/p0K4lLZo1RoKb7nmWMY9M2WK/r8ZttzYGvOjzVbRpuQMLl6ykTcsdWLx01aZtXTvtwu+vO5v+F/+epStW5/cLWd6sX7+eyy8dzvEnnES/o47eYvvxJ5zERT8ocQDnKEkBXNUY8KXAJEl/lzQ6szwNTAIuyX/36reOu7bc9Hq/zu3ZvkERny9fzcQXZzKof28aN2oAwC4778jOzZtsrZn/8Ld/zuDckw4C4NyTDuLJ56YDsGub5oy/+XsM/cn9zP5kUS1/E6srIQR+et01dOzYkYGDz9+0/uOPP9r0evLkSXTo0DFC79JByn6JrdIKOITwtKS9gF5Au8zq+cCrIYQN+e5cbGN/OZjDD+hEy2ZNmP30KEbd9RTbFRUCcPeEF/h23/04+8SDWF+6gS/Wree8K+8BYNLLs+jcoQ3Pjb0CgNVr13H+NWNZvOzfVR7z5nsnMu7GIQwa0JtPFizl3BFlbV5dchw7NWvMrVd/B4DSDRs57Jxf5+NrWx5Ne+N1nnz8MTrttRdnnNIfgGGXXs6jj0zgo4/mUlAg2rZtx7Ujfxa5p8mVpArY09Asim11GppVrjamoe195TNZZ857Nx4TNa19IYaZpUqCCmAHsJmlS0GKpqGZmSWKK2Azs0iSdBIuOTfONDPLQm1OQ5N0j6RFkt4ut24nSRMlfZD5s3lmvTK3a5gtabqk/atq3wFsZqlSUFCQ9ZKF+4BjN1t3FTAphNCJsmsirsqsPw7olFlKgN9X2dcsv5OZWSLUZgUcQngeWLrZ6v7A2MzrscCAcuvvD2VeBppJaltZ+w5gM0uV6twNTVKJpNfKLSVZHKJ1CGFB5vVC4KsbN7cD/lVuv3l8fQFbhXwSzsxSpTrn4MrftyYXIYQgKeeLzRzAZpYqdTAL4jNJbUMICzJDDF/dnGU+sGu5/dpn1m2VhyDMLFXq4GY8jwODMq8HAY+VWz8wMxviYGBFuaGKCrkCNrNUqc0r4SQ9CPQBWkqaB4yk7D7oD0saStmtes/I7P4UcDwwG1gDnL9Fg5txAJtZqtTmEEQI4aytbOpbwb4BuKg67TuAzSxVEnQhnAPYzNIlSZciO4DNLFUSlL8OYDNLF9+O0swsEg9BmJlF4gA2M4skQfnrADazdHEFbGYWSYLy1wFsZuniWRBmZpEUJKgEdgCbWaokKH8dwGaWLj4JZ2YWSYKGgB3AZpYuPglnZhaJcACbmUWRoALYAWxm6eKTcGZmkSQofx3AZpYuvhDDzCwSz4IwM4skQQWwA9jM0sVDEGZmkSQnfh3AZpYynoZmZhZJgs7BOYDNLF08C8LMLBIPQZiZRZKgAtgBbGbpkqQKuCB2B8zMapOqsVTZlnSZpHckvS3pQUkNJXWQNFXSbEkPSWqQa18dwGaWKoUFynqpjKR2wHCgZwihK1AInAncCNwSQtgTWAYMzbWvDmAzSxVJWS9ZKAIaSSoCioEFwLeACZntY4EBufbVAWxmqSJVZ1GJpNfKLSVftRNCmA/cDHxCWfCuAF4HlocQSjO7zQPa5dpXn4Qzs1Spzr0gQgijgdEVbZPUHOgPdACWA38Gjq2FLm7iADazVKnFSRD9gLkhhMVl7eovwKFAM0lFmSq4PTA/1wPkPYDn/vOWfB/CEmjEk7Nid8HqodsHdK5xG7U4De0T4GBJxcBaoC/wGjAZOA0YDwwCHsv1AB4DNrNUKZSyXioTQphK2cm2N4AZlOXlaOBK4HJJs4EWwJhc++ohCDNLldq8Ei6EMBIYudnqOUCv2mjfAWxmqeJLkc3MIknSpcgOYDNLFVfAZmaRJKgAdgCbWboUJSiBHcBmlioJyl8HsJmlix9Lb2YWSYLy1wFsZuniWRBmZpFUdaP1+sQBbGapkqD8dQCbWbooq6e91Q8OYDNLFVfAZmaROIDNzCLxzXjMzCIpTNBjJhzAZpYqvhLOzCwSjwGbmUWSoALYAWxm6VLgecBmZnG4AjYzi6QoQYPADmAzSxVXwGZmkXgamplZJAnKXwewmaVLgi6EcwCbWbp4CMLMLBIHsJlZJMmJXwewmaVMggrgRI1Xm5lVSVLWSxZtNZM0QdIsSTMl9Za0k6SJkj7I/Nk81746gM0sVQqqsWThNuDpEEJnoDswE7gKmBRC6ARMyrzPua9mZqlRIGW9VEbSjsARwBiAEMKXIYTlQH9gbGa3scCAnPua6wfNzOqj6gxBSCqR9Fq5paRcUx2AxcC9kqZJultSY6B1CGFBZp+FQOtc++qTcGaWKtWpKkMIo4HRW9lcBOwPDAshTJV0G5sNN4QQgqSQW09dAZtZytTiSbh5wLwQwtTM+wmUBfJnktpmjtUWWJRrXx3AZpYqqsZSmRDCQuBfkvbOrOoLvAs8DgzKrBsEPJZrXz0EYWapUli7E4GHAQ9IagDMAc6nrHB9WNJQ4GPgjFwbdwCbWarUZv6GEN4EelawqW9ttO8ANrNUUYIuRnYAm1mqJOlSZAewmaWKn4psZhaJK2Azs0h8P2Azs0gS9FR6B7CZpYtnQZiZRZKgEQgHcL6sW7eO4SWDWL/+SzaUbuC/+h7FkO9fzKhrr+S9me9QVFRE5y5dueLHIykq2i52d62OtGrSgMEH7rLpfcvi7Xhq1hI+WLKG73RvQ1Gh2Lgx8PBbn/HJ8i8i9jS5klQBK4Scb+STlYUr1+f3APVUCIG1a9dSXFxMael6Lv7uQIb98CpWrVzBQYccDsDPrx1B9x4HMOC0MyP3tu7d8I8PY3chOgGjjt2T3/zzI87q0ZbJs5cyc9Fq9m3dmL6dWvC7Fz6J3cU6d/uAzjVOz+ffX5p15hyx105R09oVcJ5Iori4GIDS0lJKS0uRxMGHHrFpn326fJPFiz6L1UWLbO+di1my+kuWrS0lhEDD7crujdWwqIAVa9dH7l1yJWkWhO+GlkcbNmxg6NmnMuDoI+h5UG/27dpt07bS0vU8+9QT9Op9WMQeWkz7t9+B1+etBOAvMxbRv0srfnb0Hgzo2oon3l0cuXfJVVt3Q6sLOQewpPMr2bbpLvN/vPfuXA+ReIWFhYz50yP8+W+TmPnODObM/mDTtt/+6hd073EA3XscELGHFkuhoGubJrz56SoADuvQjEffXsTIZz/k0RmLOLtH28g9TK7aeiRRnfS1Bp/92dY2hBBGhxB6hhB6nnf+d2twiHRo2nQHehzQi1deegGA+/5wJyuWL+Oiy0ZE7pnFsm/rJsxbsY5V6zYA0Gu3HXkrE8bTPl3FN5o3jNm9REtSBVzpGLCk6VvbRA2eg7QtWL5sKYVFRTRtugPrvviC1155ibMHDuHJv07glZemcMudYygo8AjQtqr88APAii9K2bNlMbOXrGGvlsUsXu0x4JzVh2TNUlUn4VoDxwDLNlsv4MW89CglPl+ymBt+eg0bN24gbAz06XcMhxzeh28d3J3Wbdpy4ZBzADj8yH4M/t4PIvfW6lKDQtG5VWMeenPhpnXjpy3k1G6tKRCs3xAYP21BJS1YZerD0EK2Kp2GJmkMcG8I4YUKtv0phHB2VQfYVqehWeU8Dc0qUhvT0F6dsyLrzDmw4471dxpaCGFoJduqDF8zszqXnALY84DNLF2SdCWcA9jMUiVBQ8AOYDNLlwTlrwPYzNJFCSqBHcBmlioJyl8HsJmlS4Ly1wFsZimToAR2AJtZqngamplZJB4DNjOLxAFsZhaJhyDMzCJJUgXsG9KaWarU9g3ZJRVKmibpycz7DpKmSpot6SFJDXLtqwPYzNKl9h+JcQkws9z7G4FbQgh7Unav9K3eNbIqDmAzS5XafCacpPbACcDdmfcCvgVMyOwyFhiQc19z/aCZWX1UnQK4/AOEM0vJZs3dCowANmbetwCWhxBKM+/nAe1y7atPwplZulTjJFwIYTQwusJmpBOBRSGE1yX1qZW+bcYBbGapUovT0A4FTpZ0PNAQ2AG4DWgmqShTBbcH5ud6AA9BmFmqSNkvlQkhXB1CaB9C2B04E/hHCOEcYDJwWma3QcBjufbVAWxmqVL7kyC2cCVwuaTZlI0Jj8m1IQ9BmFmq5OOG7CGE54DnMq/nAL1qo10HsJmlSpKuhHMAm1mqJCh/HcBmljIJSmAHsJmliu+GZmYWiceAzcwiKXAAm5nFkpwEdgCbWap4CMLMLJIE5a8D2MzSxRWwmVkk+bgUOV8cwGaWKsmJXwewmaVMggpgB7CZpYuvhDMziyU5+esANrN0SVD+OoDNLF2yedx8feEANrNUSVD++plwZmaxuAI2s1RJUgXsADazVPE0NDOzSFwBm5lF4gA2M4vEQxBmZpG4AjYziyRB+esANrOUSVACO4DNLFWSdCmyQgix+7DNkFQSQhgdux9Wv/h3se3ypch1qyR2B6xe8u9iG+UANjOLxAFsZhaJA7hueZzPKuLfxTbKJ+HMzCJxBWxmFokD2MwsEgdwNUgaLmmmpAfy1P5PJV2Rj7YtOST1kfRk7H5Y/vlKuOq5EOgXQpgXuyNmlnyugLMk6S6gI/B3SddIukfSK5KmSeqf2WewpL9KmijpI0kXS7o8s8/LknbK7Pc9Sa9KekvSI5KKKzjeHpKelvS6pP+T1Lluv7HVhKTdJc2SdJ+k9yU9IKmfpCmSPpDUK7O8lPl9vChp7wraaVzRb83SwQGcpRDCBcCnwJFAY+AfIYRemfc3SWqc2bUrcApwIHA9sCaE0AN4CRiY2ecvIYQDQwjdgZnA0AoOORoYFkI4ALgCuDM/38zyaE/gN0DnzHI2cBhl/z9/DMwCDs/8Pq4DbqigjWvY+m/NEs5DELk5Gji53HhtQ2C3zOvJIYRVwCpJK4AnMutnAN0yr7tK+gXQDGgCPFO+cUlNgEOAP+vrG4tsn48vYnk1N4QwA0DSO8CkEEKQNAPYHdgRGCupExCA7SpoY2u/tZn57rzlnwM4NwJODSG89x8rpYOAdeVWbSz3fiNf//e+DxgQQnhL0mCgz2btFwDLQwj71W63rY5V9VsYRdlf2N+WtDvwXAVtVPhbs3TwEERungGGKVOeSupRzc83BRZI2g44Z/ONIYSVwFxJp2fal6TuNeyz1T87AvMzrwdvZZ+a/tasHnMA52YUZf9cnJ75p+Woan7+J8BUYApl44AVOQcYKukt4B3AJ1/S59fALyVNY+v/Gq3pb83qMV+KbGYWiStgM7NIHMBmZpE4gM3MInEAm5lF4gA2M4vEAWxmFokD2Mwskv8HPDrvN8Li5CEAAAAASUVORK5CYII=\n",
            "text/plain": [
              "<Figure size 432x288 with 2 Axes>"
            ]
          },
          "metadata": {
            "tags": [],
            "needs_background": "light"
          }
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "fqeihwLm5bb9"
      },
      "source": [
        "Comparing with SKlearn's Logistic regression"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "T-YIX4xU5F6_",
        "outputId": "92b1bbcd-5b8c-4692-dbf9-ab838572f836"
      },
      "source": [
        "clf = LogisticRegression(random_state=1,max_iter=100).fit(X_train_std, y_train)\n",
        "sk_score = clf.score(X_test_std, y_test)\n",
        "print(sk_score)\n",
        "\n",
        "clf = LogisticRegression(random_state=1,max_iter=1000).fit(X_train_org, y_train)\n",
        "sk_score = clf.score(X_test_org, y_test)\n",
        "print(sk_score)"
      ],
      "execution_count": 47,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "0.7940074906367042\n",
            "0.7902621722846442\n"
          ],
          "name": "stdout"
        }
      ]
    }
  ]
}