Thyroid_Prediction .ipynb

{
  "nbformat": 4,
  "nbformat_minor": 0,
  "metadata": {
    "colab": {
      "name": "Untitled9.ipynb",
      "provenance": []
    },
    "kernelspec": {
      "name": "python3",
      "display_name": "Python 3"
    },
    "language_info": {
      "name": "python"
    }
  },
  "cells": [
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "J9dl6BEDU_Ys",
        "outputId": "de9a372a-22b9-4737-84dc-a9537f2464ac"
      },
      "source": [
        "from google.colab import drive\n",
        "drive.mount('/content/drive')"
      ],
      "execution_count": 27,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "Drive already mounted at /content/drive; to attempt to forcibly remount, call drive.mount(\"/content/drive\", force_remount=True).\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "AYKTJkRBRNpZ"
      },
      "source": [
        "**Diagnosing Hypothyroid disease using Deep Learning**\n",
        "\n",
        "Hypothyroid disease or Hypothyroidism is one of the most difficult diseases to diagnose. It is a condition which causes your thyroid gland to not produce enough of certain crucial hormones. It usually may not cause noticeable symptoms in the early stages and hence becomes extremely difficult to diagnose. Applying deep learning algorithms can help assist healthcare workers in identifying signs and symptoms based on patterns shown in past cases of patients' data.\n"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "UrfQa5jhrpZn"
      },
      "source": [
        "import numpy as np # linear algebra\n",
        "import pandas as pd # data processing, CSV file I/O (e.g. pd.read_csv)\n",
        "import pickle # For saving models\n",
        "import matplotlib.pyplot as plt # For plotting figures\n",
        "from sklearn.metrics import confusion_matrix, f1_score, precision_score, recall_score\n",
        "\n",
        "# For building our model\n",
        "import keras\n",
        "from keras.models import Sequential\n",
        "from keras.layers import Dense"
      ],
      "execution_count": 28,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "VgmDaCL_SfQM"
      },
      "source": [
        "**Data**\n",
        "\n",
        "The dataset contains the value of different hormone levels of different patients along with our target variable, i.e., if the patient was suffering from hypothyroidism or not.\n"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 232
        },
        "id": "6KnA0FedSYBd",
        "outputId": "ce291146-6f0e-4dd7-b978-f7f23ca72541"
      },
      "source": [
        "# Load dataset from csv using pandas\n",
        "dataset = pd.read_csv('/content/drive/MyDrive/hypothyroid.csv')\n",
        "dataset.head()"
      ],
      "execution_count": 29,
      "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>Unnamed: 0</th>\n",
              "      <th>Age</th>\n",
              "      <th>Sex</th>\n",
              "      <th>on_thyroxine</th>\n",
              "      <th>query_on_thyroxine</th>\n",
              "      <th>on_antithyroid_medication</th>\n",
              "      <th>thyroid_surgery</th>\n",
              "      <th>query_hypothyroid</th>\n",
              "      <th>query_hyperthyroid</th>\n",
              "      <th>pregnant</th>\n",
              "      <th>sick</th>\n",
              "      <th>tumor</th>\n",
              "      <th>lithium</th>\n",
              "      <th>goitre</th>\n",
              "      <th>TSH_measured</th>\n",
              "      <th>TSH</th>\n",
              "      <th>T3_measured</th>\n",
              "      <th>T3</th>\n",
              "      <th>TT4_measured</th>\n",
              "      <th>TT4</th>\n",
              "      <th>T4U_measured</th>\n",
              "      <th>T4U</th>\n",
              "      <th>FTI_measured</th>\n",
              "      <th>FTI</th>\n",
              "      <th>TBG_measured</th>\n",
              "      <th>TBG</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>0</th>\n",
              "      <td>hypothyroid</td>\n",
              "      <td>72</td>\n",
              "      <td>M</td>\n",
              "      <td>f</td>\n",
              "      <td>f</td>\n",
              "      <td>f</td>\n",
              "      <td>f</td>\n",
              "      <td>f</td>\n",
              "      <td>f</td>\n",
              "      <td>f</td>\n",
              "      <td>f</td>\n",
              "      <td>f</td>\n",
              "      <td>f</td>\n",
              "      <td>f</td>\n",
              "      <td>y</td>\n",
              "      <td>30</td>\n",
              "      <td>y</td>\n",
              "      <td>0.6</td>\n",
              "      <td>y</td>\n",
              "      <td>15</td>\n",
              "      <td>y</td>\n",
              "      <td>1.48</td>\n",
              "      <td>y</td>\n",
              "      <td>10</td>\n",
              "      <td>n</td>\n",
              "      <td>?</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1</th>\n",
              "      <td>hypothyroid</td>\n",
              "      <td>15</td>\n",
              "      <td>F</td>\n",
              "      <td>t</td>\n",
              "      <td>f</td>\n",
              "      <td>f</td>\n",
              "      <td>f</td>\n",
              "      <td>f</td>\n",
              "      <td>f</td>\n",
              "      <td>f</td>\n",
              "      <td>f</td>\n",
              "      <td>f</td>\n",
              "      <td>f</td>\n",
              "      <td>f</td>\n",
              "      <td>y</td>\n",
              "      <td>145</td>\n",
              "      <td>y</td>\n",
              "      <td>1.7</td>\n",
              "      <td>y</td>\n",
              "      <td>19</td>\n",
              "      <td>y</td>\n",
              "      <td>1.13</td>\n",
              "      <td>y</td>\n",
              "      <td>17</td>\n",
              "      <td>n</td>\n",
              "      <td>?</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2</th>\n",
              "      <td>hypothyroid</td>\n",
              "      <td>24</td>\n",
              "      <td>M</td>\n",
              "      <td>f</td>\n",
              "      <td>f</td>\n",
              "      <td>f</td>\n",
              "      <td>f</td>\n",
              "      <td>f</td>\n",
              "      <td>f</td>\n",
              "      <td>f</td>\n",
              "      <td>f</td>\n",
              "      <td>f</td>\n",
              "      <td>f</td>\n",
              "      <td>f</td>\n",
              "      <td>y</td>\n",
              "      <td>0</td>\n",
              "      <td>y</td>\n",
              "      <td>0.2</td>\n",
              "      <td>y</td>\n",
              "      <td>4</td>\n",
              "      <td>y</td>\n",
              "      <td>1</td>\n",
              "      <td>y</td>\n",
              "      <td>0</td>\n",
              "      <td>n</td>\n",
              "      <td>?</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>3</th>\n",
              "      <td>hypothyroid</td>\n",
              "      <td>24</td>\n",
              "      <td>F</td>\n",
              "      <td>f</td>\n",
              "      <td>f</td>\n",
              "      <td>f</td>\n",
              "      <td>f</td>\n",
              "      <td>f</td>\n",
              "      <td>f</td>\n",
              "      <td>f</td>\n",
              "      <td>f</td>\n",
              "      <td>f</td>\n",
              "      <td>f</td>\n",
              "      <td>f</td>\n",
              "      <td>y</td>\n",
              "      <td>430</td>\n",
              "      <td>y</td>\n",
              "      <td>0.4</td>\n",
              "      <td>y</td>\n",
              "      <td>6</td>\n",
              "      <td>y</td>\n",
              "      <td>1.04</td>\n",
              "      <td>y</td>\n",
              "      <td>6</td>\n",
              "      <td>n</td>\n",
              "      <td>?</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>4</th>\n",
              "      <td>hypothyroid</td>\n",
              "      <td>77</td>\n",
              "      <td>M</td>\n",
              "      <td>f</td>\n",
              "      <td>f</td>\n",
              "      <td>f</td>\n",
              "      <td>f</td>\n",
              "      <td>f</td>\n",
              "      <td>f</td>\n",
              "      <td>f</td>\n",
              "      <td>f</td>\n",
              "      <td>f</td>\n",
              "      <td>f</td>\n",
              "      <td>f</td>\n",
              "      <td>y</td>\n",
              "      <td>7.3</td>\n",
              "      <td>y</td>\n",
              "      <td>1.2</td>\n",
              "      <td>y</td>\n",
              "      <td>57</td>\n",
              "      <td>y</td>\n",
              "      <td>1.28</td>\n",
              "      <td>y</td>\n",
              "      <td>44</td>\n",
              "      <td>n</td>\n",
              "      <td>?</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>"
            ],
            "text/plain": [
              "    Unnamed: 0 Age Sex on_thyroxine  ... FTI_measured FTI TBG_measured TBG\n",
              "0  hypothyroid  72   M            f  ...            y  10            n   ?\n",
              "1  hypothyroid  15   F            t  ...            y  17            n   ?\n",
              "2  hypothyroid  24   M            f  ...            y   0            n   ?\n",
              "3  hypothyroid  24   F            f  ...            y   6            n   ?\n",
              "4  hypothyroid  77   M            f  ...            y  44            n   ?\n",
              "\n",
              "[5 rows x 26 columns]"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 29
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "aBKP4DhrWOoZ"
      },
      "source": [
        "**Data Cleaning**\n",
        "\n"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "BLyXpyOVSvvo"
      },
      "source": [
        "# Renaming the first column as target\n",
        "dataset = dataset.rename(columns = {dataset.columns[0]:\"target\"})\n"
      ],
      "execution_count": 30,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "dNPzGbmFU6AC",
        "outputId": "0698c062-edf0-416e-eed8-d2dcfb0a6ba9"
      },
      "source": [
        "# Check the count of data in target\n",
        "dataset[\"target\"].value_counts()"
      ],
      "execution_count": 31,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "negative       3012\n",
              "hypothyroid     151\n",
              "Name: target, dtype: int64"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 31
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "RB4S3-4bWo-8"
      },
      "source": [
        "The value 'hypothyroid' means that the patient was diagnosed with hypothyroidism and ''negative'' means that there hypothyroidism was not diagnosed. We will be mapping the value ''hypothyroid'' as '1' and ''negative'' as '0'."
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "9AIwRGPBWoxb"
      },
      "source": [
        "dataset[\"target\"] = dataset[\"target\"].map({\"negative\":0,\"hypothyroid\":1})"
      ],
      "execution_count": 32,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "e92WZ6jNWnyl",
        "outputId": "c01f4606-d29d-456d-eda6-b4ac22062161"
      },
      "source": [
        "print(\"Unique categories in the column 'pregnant'\", dataset['pregnant'].unique())\n",
        "print(\"Count of categories in the column 'pregnant' \\n\", dataset[\"pregnant\"].value_counts())"
      ],
      "execution_count": 33,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "Unique categories in the column 'pregnant' ['f' 't']\n",
            "Count of categories in the column 'pregnant' \n",
            " f    3100\n",
            "t      63\n",
            "Name: pregnant, dtype: int64\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "ZatA5MHYU7Sj",
        "outputId": "84630651-eff5-4ca5-a213-9546e37044c9"
      },
      "source": [
        "print(\"Unique categories in the column 'T3 measured'\", dataset['T3_measured'].unique())\n",
        "print(\"Count of categories in the column 'T3 measured' \\n\", dataset[\"T3_measured\"].value_counts())"
      ],
      "execution_count": 34,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "Unique categories in the column 'T3 measured' ['y' 'n']\n",
            "Count of categories in the column 'T3 measured' \n",
            " y    2468\n",
            "n     695\n",
            "Name: T3_measured, dtype: int64\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "Kx-n4jWOW9g0",
        "outputId": "23914f57-ed89-43f4-a857-29c95e0ad4f7"
      },
      "source": [
        "print(\"Unique categories in the column 'Sex'\", dataset['Sex'].unique())\n",
        "print(\"Count of categories in the column 'Sex' \\n\", dataset[\"Sex\"].value_counts())"
      ],
      "execution_count": 35,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "Unique categories in the column 'Sex' ['M' 'F' '?']\n",
            "Count of categories in the column 'Sex' \n",
            " F    2182\n",
            "M     908\n",
            "?      73\n",
            "Name: Sex, dtype: int64\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 215
        },
        "id": "6utobRh1XB4i",
        "outputId": "c8396093-35de-4fb6-df4b-a3887ba82bcb"
      },
      "source": [
        "# Replacing the categorical values into binary values\n",
        "dataset = dataset.replace({'f':0,'t':1, 'y':1, 'n':0, 'M':0, 'F':1})\n",
        "# Displaying the head of the dataset\n",
        "dataset.head()"
      ],
      "execution_count": 36,
      "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>target</th>\n",
              "      <th>Age</th>\n",
              "      <th>Sex</th>\n",
              "      <th>on_thyroxine</th>\n",
              "      <th>query_on_thyroxine</th>\n",
              "      <th>on_antithyroid_medication</th>\n",
              "      <th>thyroid_surgery</th>\n",
              "      <th>query_hypothyroid</th>\n",
              "      <th>query_hyperthyroid</th>\n",
              "      <th>pregnant</th>\n",
              "      <th>sick</th>\n",
              "      <th>tumor</th>\n",
              "      <th>lithium</th>\n",
              "      <th>goitre</th>\n",
              "      <th>TSH_measured</th>\n",
              "      <th>TSH</th>\n",
              "      <th>T3_measured</th>\n",
              "      <th>T3</th>\n",
              "      <th>TT4_measured</th>\n",
              "      <th>TT4</th>\n",
              "      <th>T4U_measured</th>\n",
              "      <th>T4U</th>\n",
              "      <th>FTI_measured</th>\n",
              "      <th>FTI</th>\n",
              "      <th>TBG_measured</th>\n",
              "      <th>TBG</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>0</th>\n",
              "      <td>1</td>\n",
              "      <td>72</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>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>1</td>\n",
              "      <td>30</td>\n",
              "      <td>1</td>\n",
              "      <td>0.6</td>\n",
              "      <td>1</td>\n",
              "      <td>15</td>\n",
              "      <td>1</td>\n",
              "      <td>1.48</td>\n",
              "      <td>1</td>\n",
              "      <td>10</td>\n",
              "      <td>0</td>\n",
              "      <td>?</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1</th>\n",
              "      <td>1</td>\n",
              "      <td>15</td>\n",
              "      <td>1</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>0</td>\n",
              "      <td>0</td>\n",
              "      <td>1</td>\n",
              "      <td>145</td>\n",
              "      <td>1</td>\n",
              "      <td>1.7</td>\n",
              "      <td>1</td>\n",
              "      <td>19</td>\n",
              "      <td>1</td>\n",
              "      <td>1.13</td>\n",
              "      <td>1</td>\n",
              "      <td>17</td>\n",
              "      <td>0</td>\n",
              "      <td>?</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2</th>\n",
              "      <td>1</td>\n",
              "      <td>24</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>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>1</td>\n",
              "      <td>0.2</td>\n",
              "      <td>1</td>\n",
              "      <td>4</td>\n",
              "      <td>1</td>\n",
              "      <td>1</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>?</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>3</th>\n",
              "      <td>1</td>\n",
              "      <td>24</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>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>1</td>\n",
              "      <td>430</td>\n",
              "      <td>1</td>\n",
              "      <td>0.4</td>\n",
              "      <td>1</td>\n",
              "      <td>6</td>\n",
              "      <td>1</td>\n",
              "      <td>1.04</td>\n",
              "      <td>1</td>\n",
              "      <td>6</td>\n",
              "      <td>0</td>\n",
              "      <td>?</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>4</th>\n",
              "      <td>1</td>\n",
              "      <td>77</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>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>1</td>\n",
              "      <td>7.3</td>\n",
              "      <td>1</td>\n",
              "      <td>1.2</td>\n",
              "      <td>1</td>\n",
              "      <td>57</td>\n",
              "      <td>1</td>\n",
              "      <td>1.28</td>\n",
              "      <td>1</td>\n",
              "      <td>44</td>\n",
              "      <td>0</td>\n",
              "      <td>?</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>"
            ],
            "text/plain": [
              "   target Age Sex  on_thyroxine  ...  FTI_measured  FTI  TBG_measured  TBG\n",
              "0       1  72   0             0  ...             1   10             0    ?\n",
              "1       1  15   1             1  ...             1   17             0    ?\n",
              "2       1  24   0             0  ...             1    0             0    ?\n",
              "3       1  24   1             0  ...             1    6             0    ?\n",
              "4       1  77   0             0  ...             1   44             0    ?\n",
              "\n",
              "[5 rows x 26 columns]"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 36
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "UkbMnjfUXOAe"
      },
      "source": [
        "Some of the features contain '?' values. So we will be replacing them with NaN values to count them."
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "S0sYKSwPXFvO"
      },
      "source": [
        "# Replacing ? into NaN values \n",
        "dataset.replace(to_replace='?', inplace=True, value=np.NaN)"
      ],
      "execution_count": 37,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "agNL15c5XXl5"
      },
      "source": [
        "Checking for null values\n"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "n8IhbbIeXRl0",
        "outputId": "84af7e97-a0b1-4be5-d0a3-be8c5282d271"
      },
      "source": [
        "# Count the number of null values\n",
        "dataset.isnull().sum()"
      ],
      "execution_count": 38,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "target                          0\n",
              "Age                           446\n",
              "Sex                            73\n",
              "on_thyroxine                    0\n",
              "query_on_thyroxine              0\n",
              "on_antithyroid_medication       0\n",
              "thyroid_surgery                 0\n",
              "query_hypothyroid               0\n",
              "query_hyperthyroid              0\n",
              "pregnant                        0\n",
              "sick                            0\n",
              "tumor                           0\n",
              "lithium                         0\n",
              "goitre                          0\n",
              "TSH_measured                    0\n",
              "TSH                           468\n",
              "T3_measured                     0\n",
              "T3                            695\n",
              "TT4_measured                    0\n",
              "TT4                           249\n",
              "T4U_measured                    0\n",
              "T4U                           248\n",
              "FTI_measured                    0\n",
              "FTI                           247\n",
              "TBG_measured                    0\n",
              "TBG                          2903\n",
              "dtype: int64"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 38
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "wn7ddKBbX59h"
      },
      "source": [
        "Here, we can see that the feature column \"TBG\" contains extremely high number of null values. So, we will be not be using this column for our model."
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "KTpU-YZDX0qv"
      },
      "source": [
        "# Dropping the TBG column as it contains extremely high number of null values\n",
        "dataset.drop('TBG', axis = 1, inplace=True)"
      ],
      "execution_count": 39,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "vtB6z-8MX-JO"
      },
      "source": [
        "# Selecting columns with data type as 'object'\n",
        "columns = dataset.columns[dataset.dtypes.eq('object')]\n",
        "\n",
        "# Convert to numeric values\n",
        "dataset[columns] = dataset[columns].apply(pd.to_numeric, errors='coerce')"
      ],
      "execution_count": 40,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "0eUq_JVXYKjK"
      },
      "source": [
        "**Visualize the patterns or trends in data**"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 1000
        },
        "id": "ZoYzUt3qYJ4U",
        "outputId": "37a64f38-2c0f-4979-96fe-11648e9ac940"
      },
      "source": [
        "dataset.hist(figsize = (20,20));"
      ],
      "execution_count": 41,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<Figure size 1440x1440 with 25 Axes>"
            ]
          },
          "metadata": {
            "tags": [],
            "needs_background": "light"
          }
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "hVfPh4QwYfVA"
      },
      "source": [
        "**Replacing null values in our data**"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "Rn09bFUcYDyC",
        "outputId": "2e9372fc-6fad-4b0e-de19-bc6c6232d0e5"
      },
      "source": [
        "# List of columns containing null values\n",
        "dataset.columns[dataset.isnull().any()].tolist() "
      ],
      "execution_count": 42,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "['Age', 'Sex', 'TSH', 'T3', 'TT4', 'T4U', 'FTI']"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 42
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "ZKhBzjP0Ym_d"
      },
      "source": [
        "# Replacing null values by mean \n",
        "dataset['Age'].fillna(dataset['Age'].mean(), inplace = True)\n",
        "dataset['T4U'].fillna(dataset['T4U'].mean(), inplace = True)\n",
        "\n",
        "# Replacing null values by median\n",
        "dataset['TSH'].fillna(dataset['TSH'].mean(), inplace = True)\n",
        "dataset['T3'].fillna(dataset['T3'].median(), inplace = True)\n",
        "dataset['TT4'].fillna(dataset['TT4'].median(), inplace = True)\n",
        "dataset['FTI'].fillna(dataset['FTI'].median(), inplace = True)\n",
        "\n",
        "# The gender data looks to be imbalanced with 0 lesser than 1\n",
        "# Replacing null values with 0\n",
        "dataset['Sex'].fillna(0, inplace = True)"
      ],
      "execution_count": 43,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "yLrxBsEXY2Yr"
      },
      "source": [
        "**Modeling the data**\n",
        "\n",
        "We will be using a Neural network for predicting the hypothyroidism (target) based on our input data (features)."
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "Lh6SMvhYYvTQ"
      },
      "source": [
        "# Features\n",
        "X = dataset.drop('target', axis = 1) # selecting all columns except the target\n",
        "# Target variable\n",
        "y = dataset['target']\n"
      ],
      "execution_count": 44,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "hEZJhZUVZDeG"
      },
      "source": [
        "from sklearn.model_selection import train_test_split\n",
        "\n",
        "# Split the data\n",
        "X_train, X_test, y_train, y_test = train_test_split(X, y, test_size=0.2, random_state=42)"
      ],
      "execution_count": 45,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "eKkcgWNGZJUi"
      },
      "source": [
        "from sklearn.preprocessing import StandardScaler\n",
        "\n",
        "# Initialization of the class\n",
        "scaler = StandardScaler()\n",
        "\n",
        "# Applying the scaler on test and train data\n",
        "X_train = scaler.fit_transform(X_train)\n",
        "X_test = scaler.transform(X_test)"
      ],
      "execution_count": 46,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "7mB8u3lUZNG7"
      },
      "source": [
        "# Input\n",
        "model = Sequential()\n",
        "# Hidden layer\n",
        "model.add(Dense(64, kernel_initializer='uniform', input_dim=24, activation='relu'))\n",
        "# Output layer\n",
        "model.add(Dense(1, kernel_initializer='uniform', activation='sigmoid'))"
      ],
      "execution_count": 47,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "o97bSFXmZQ8P"
      },
      "source": [
        "model.compile(loss='binary_crossentropy', optimizer='adam', metrics=['accuracy'])"
      ],
      "execution_count": 48,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "1xy8yvzZZUIl",
        "outputId": "3eed9ae0-4372-42ef-bd41-e88544adacfd"
      },
      "source": [
        "# Training the model\n",
        "result = model.fit(X_train, y_train, epochs=100, validation_split=0.2, batch_size=40,  verbose=2)\n"
      ],
      "execution_count": 49,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "Epoch 1/100\n",
            "51/51 - 1s - loss: 0.5803 - accuracy: 0.9521 - val_loss: 0.4277 - val_accuracy: 0.9625\n",
            "Epoch 2/100\n",
            "51/51 - 0s - loss: 0.2917 - accuracy: 0.9555 - val_loss: 0.1714 - val_accuracy: 0.9644\n",
            "Epoch 3/100\n",
            "51/51 - 0s - loss: 0.1375 - accuracy: 0.9644 - val_loss: 0.0961 - val_accuracy: 0.9743\n",
            "Epoch 4/100\n",
            "51/51 - 0s - loss: 0.0919 - accuracy: 0.9694 - val_loss: 0.0708 - val_accuracy: 0.9842\n",
            "Epoch 5/100\n",
            "51/51 - 0s - loss: 0.0723 - accuracy: 0.9768 - val_loss: 0.0589 - val_accuracy: 0.9842\n",
            "Epoch 6/100\n",
            "51/51 - 0s - loss: 0.0625 - accuracy: 0.9792 - val_loss: 0.0528 - val_accuracy: 0.9842\n",
            "Epoch 7/100\n",
            "51/51 - 0s - loss: 0.0570 - accuracy: 0.9792 - val_loss: 0.0498 - val_accuracy: 0.9842\n",
            "Epoch 8/100\n",
            "51/51 - 0s - loss: 0.0535 - accuracy: 0.9802 - val_loss: 0.0478 - val_accuracy: 0.9822\n",
            "Epoch 9/100\n",
            "51/51 - 0s - loss: 0.0508 - accuracy: 0.9817 - val_loss: 0.0452 - val_accuracy: 0.9822\n",
            "Epoch 10/100\n",
            "51/51 - 0s - loss: 0.0491 - accuracy: 0.9822 - val_loss: 0.0441 - val_accuracy: 0.9802\n",
            "Epoch 11/100\n",
            "51/51 - 0s - loss: 0.0471 - accuracy: 0.9817 - val_loss: 0.0440 - val_accuracy: 0.9822\n",
            "Epoch 12/100\n",
            "51/51 - 0s - loss: 0.0457 - accuracy: 0.9822 - val_loss: 0.0427 - val_accuracy: 0.9822\n",
            "Epoch 13/100\n",
            "51/51 - 0s - loss: 0.0447 - accuracy: 0.9832 - val_loss: 0.0421 - val_accuracy: 0.9842\n",
            "Epoch 14/100\n",
            "51/51 - 0s - loss: 0.0432 - accuracy: 0.9852 - val_loss: 0.0414 - val_accuracy: 0.9842\n",
            "Epoch 15/100\n",
            "51/51 - 0s - loss: 0.0424 - accuracy: 0.9837 - val_loss: 0.0412 - val_accuracy: 0.9822\n",
            "Epoch 16/100\n",
            "51/51 - 0s - loss: 0.0412 - accuracy: 0.9867 - val_loss: 0.0397 - val_accuracy: 0.9822\n",
            "Epoch 17/100\n",
            "51/51 - 0s - loss: 0.0405 - accuracy: 0.9867 - val_loss: 0.0387 - val_accuracy: 0.9842\n",
            "Epoch 18/100\n",
            "51/51 - 0s - loss: 0.0398 - accuracy: 0.9867 - val_loss: 0.0392 - val_accuracy: 0.9842\n",
            "Epoch 19/100\n",
            "51/51 - 0s - loss: 0.0393 - accuracy: 0.9867 - val_loss: 0.0385 - val_accuracy: 0.9822\n",
            "Epoch 20/100\n",
            "51/51 - 0s - loss: 0.0384 - accuracy: 0.9867 - val_loss: 0.0378 - val_accuracy: 0.9842\n",
            "Epoch 21/100\n",
            "51/51 - 0s - loss: 0.0377 - accuracy: 0.9862 - val_loss: 0.0383 - val_accuracy: 0.9842\n",
            "Epoch 22/100\n",
            "51/51 - 0s - loss: 0.0372 - accuracy: 0.9876 - val_loss: 0.0371 - val_accuracy: 0.9822\n",
            "Epoch 23/100\n",
            "51/51 - 0s - loss: 0.0363 - accuracy: 0.9862 - val_loss: 0.0372 - val_accuracy: 0.9842\n",
            "Epoch 24/100\n",
            "51/51 - 0s - loss: 0.0360 - accuracy: 0.9872 - val_loss: 0.0369 - val_accuracy: 0.9842\n",
            "Epoch 25/100\n",
            "51/51 - 0s - loss: 0.0355 - accuracy: 0.9867 - val_loss: 0.0369 - val_accuracy: 0.9842\n",
            "Epoch 26/100\n",
            "51/51 - 0s - loss: 0.0350 - accuracy: 0.9876 - val_loss: 0.0359 - val_accuracy: 0.9862\n",
            "Epoch 27/100\n",
            "51/51 - 0s - loss: 0.0354 - accuracy: 0.9862 - val_loss: 0.0352 - val_accuracy: 0.9862\n",
            "Epoch 28/100\n",
            "51/51 - 0s - loss: 0.0343 - accuracy: 0.9872 - val_loss: 0.0355 - val_accuracy: 0.9862\n",
            "Epoch 29/100\n",
            "51/51 - 0s - loss: 0.0339 - accuracy: 0.9867 - val_loss: 0.0353 - val_accuracy: 0.9862\n",
            "Epoch 30/100\n",
            "51/51 - 0s - loss: 0.0333 - accuracy: 0.9876 - val_loss: 0.0360 - val_accuracy: 0.9842\n",
            "Epoch 31/100\n",
            "51/51 - 0s - loss: 0.0330 - accuracy: 0.9876 - val_loss: 0.0351 - val_accuracy: 0.9862\n",
            "Epoch 32/100\n",
            "51/51 - 0s - loss: 0.0325 - accuracy: 0.9876 - val_loss: 0.0358 - val_accuracy: 0.9842\n",
            "Epoch 33/100\n",
            "51/51 - 0s - loss: 0.0324 - accuracy: 0.9881 - val_loss: 0.0349 - val_accuracy: 0.9862\n",
            "Epoch 34/100\n",
            "51/51 - 0s - loss: 0.0323 - accuracy: 0.9881 - val_loss: 0.0346 - val_accuracy: 0.9862\n",
            "Epoch 35/100\n",
            "51/51 - 0s - loss: 0.0315 - accuracy: 0.9886 - val_loss: 0.0353 - val_accuracy: 0.9862\n",
            "Epoch 36/100\n",
            "51/51 - 0s - loss: 0.0315 - accuracy: 0.9896 - val_loss: 0.0343 - val_accuracy: 0.9862\n",
            "Epoch 37/100\n",
            "51/51 - 0s - loss: 0.0313 - accuracy: 0.9886 - val_loss: 0.0346 - val_accuracy: 0.9842\n",
            "Epoch 38/100\n",
            "51/51 - 0s - loss: 0.0306 - accuracy: 0.9886 - val_loss: 0.0356 - val_accuracy: 0.9862\n",
            "Epoch 39/100\n",
            "51/51 - 0s - loss: 0.0309 - accuracy: 0.9886 - val_loss: 0.0360 - val_accuracy: 0.9862\n",
            "Epoch 40/100\n",
            "51/51 - 0s - loss: 0.0303 - accuracy: 0.9901 - val_loss: 0.0350 - val_accuracy: 0.9862\n",
            "Epoch 41/100\n",
            "51/51 - 0s - loss: 0.0300 - accuracy: 0.9891 - val_loss: 0.0352 - val_accuracy: 0.9862\n",
            "Epoch 42/100\n",
            "51/51 - 0s - loss: 0.0297 - accuracy: 0.9906 - val_loss: 0.0352 - val_accuracy: 0.9862\n",
            "Epoch 43/100\n",
            "51/51 - 0s - loss: 0.0293 - accuracy: 0.9901 - val_loss: 0.0350 - val_accuracy: 0.9862\n",
            "Epoch 44/100\n",
            "51/51 - 0s - loss: 0.0291 - accuracy: 0.9886 - val_loss: 0.0352 - val_accuracy: 0.9862\n",
            "Epoch 45/100\n",
            "51/51 - 0s - loss: 0.0290 - accuracy: 0.9901 - val_loss: 0.0353 - val_accuracy: 0.9862\n",
            "Epoch 46/100\n",
            "51/51 - 0s - loss: 0.0286 - accuracy: 0.9901 - val_loss: 0.0345 - val_accuracy: 0.9881\n",
            "Epoch 47/100\n",
            "51/51 - 0s - loss: 0.0285 - accuracy: 0.9906 - val_loss: 0.0353 - val_accuracy: 0.9862\n",
            "Epoch 48/100\n",
            "51/51 - 0s - loss: 0.0282 - accuracy: 0.9906 - val_loss: 0.0350 - val_accuracy: 0.9881\n",
            "Epoch 49/100\n",
            "51/51 - 0s - loss: 0.0279 - accuracy: 0.9906 - val_loss: 0.0356 - val_accuracy: 0.9862\n",
            "Epoch 50/100\n",
            "51/51 - 0s - loss: 0.0278 - accuracy: 0.9901 - val_loss: 0.0352 - val_accuracy: 0.9881\n",
            "Epoch 51/100\n",
            "51/51 - 0s - loss: 0.0278 - accuracy: 0.9921 - val_loss: 0.0343 - val_accuracy: 0.9842\n",
            "Epoch 52/100\n",
            "51/51 - 0s - loss: 0.0276 - accuracy: 0.9911 - val_loss: 0.0349 - val_accuracy: 0.9862\n",
            "Epoch 53/100\n",
            "51/51 - 0s - loss: 0.0270 - accuracy: 0.9916 - val_loss: 0.0348 - val_accuracy: 0.9881\n",
            "Epoch 54/100\n",
            "51/51 - 0s - loss: 0.0272 - accuracy: 0.9906 - val_loss: 0.0359 - val_accuracy: 0.9862\n",
            "Epoch 55/100\n",
            "51/51 - 0s - loss: 0.0267 - accuracy: 0.9911 - val_loss: 0.0352 - val_accuracy: 0.9881\n",
            "Epoch 56/100\n",
            "51/51 - 0s - loss: 0.0267 - accuracy: 0.9906 - val_loss: 0.0349 - val_accuracy: 0.9881\n",
            "Epoch 57/100\n",
            "51/51 - 0s - loss: 0.0261 - accuracy: 0.9911 - val_loss: 0.0350 - val_accuracy: 0.9881\n",
            "Epoch 58/100\n",
            "51/51 - 0s - loss: 0.0259 - accuracy: 0.9916 - val_loss: 0.0350 - val_accuracy: 0.9881\n",
            "Epoch 59/100\n",
            "51/51 - 0s - loss: 0.0257 - accuracy: 0.9916 - val_loss: 0.0350 - val_accuracy: 0.9881\n",
            "Epoch 60/100\n",
            "51/51 - 0s - loss: 0.0259 - accuracy: 0.9921 - val_loss: 0.0348 - val_accuracy: 0.9862\n",
            "Epoch 61/100\n",
            "51/51 - 0s - loss: 0.0256 - accuracy: 0.9916 - val_loss: 0.0351 - val_accuracy: 0.9881\n",
            "Epoch 62/100\n",
            "51/51 - 0s - loss: 0.0252 - accuracy: 0.9926 - val_loss: 0.0354 - val_accuracy: 0.9881\n",
            "Epoch 63/100\n",
            "51/51 - 0s - loss: 0.0250 - accuracy: 0.9911 - val_loss: 0.0359 - val_accuracy: 0.9901\n",
            "Epoch 64/100\n",
            "51/51 - 0s - loss: 0.0250 - accuracy: 0.9911 - val_loss: 0.0355 - val_accuracy: 0.9862\n",
            "Epoch 65/100\n",
            "51/51 - 0s - loss: 0.0253 - accuracy: 0.9921 - val_loss: 0.0358 - val_accuracy: 0.9862\n",
            "Epoch 66/100\n",
            "51/51 - 0s - loss: 0.0249 - accuracy: 0.9921 - val_loss: 0.0362 - val_accuracy: 0.9881\n",
            "Epoch 67/100\n",
            "51/51 - 0s - loss: 0.0245 - accuracy: 0.9926 - val_loss: 0.0356 - val_accuracy: 0.9862\n",
            "Epoch 68/100\n",
            "51/51 - 0s - loss: 0.0242 - accuracy: 0.9921 - val_loss: 0.0362 - val_accuracy: 0.9862\n",
            "Epoch 69/100\n",
            "51/51 - 0s - loss: 0.0240 - accuracy: 0.9931 - val_loss: 0.0355 - val_accuracy: 0.9862\n",
            "Epoch 70/100\n",
            "51/51 - 0s - loss: 0.0242 - accuracy: 0.9931 - val_loss: 0.0358 - val_accuracy: 0.9862\n",
            "Epoch 71/100\n",
            "51/51 - 0s - loss: 0.0239 - accuracy: 0.9931 - val_loss: 0.0355 - val_accuracy: 0.9862\n",
            "Epoch 72/100\n",
            "51/51 - 0s - loss: 0.0238 - accuracy: 0.9926 - val_loss: 0.0366 - val_accuracy: 0.9862\n",
            "Epoch 73/100\n",
            "51/51 - 0s - loss: 0.0235 - accuracy: 0.9926 - val_loss: 0.0363 - val_accuracy: 0.9862\n",
            "Epoch 74/100\n",
            "51/51 - 0s - loss: 0.0235 - accuracy: 0.9926 - val_loss: 0.0365 - val_accuracy: 0.9881\n",
            "Epoch 75/100\n",
            "51/51 - 0s - loss: 0.0232 - accuracy: 0.9931 - val_loss: 0.0367 - val_accuracy: 0.9862\n",
            "Epoch 76/100\n",
            "51/51 - 0s - loss: 0.0230 - accuracy: 0.9931 - val_loss: 0.0369 - val_accuracy: 0.9862\n",
            "Epoch 77/100\n",
            "51/51 - 0s - loss: 0.0230 - accuracy: 0.9926 - val_loss: 0.0370 - val_accuracy: 0.9862\n",
            "Epoch 78/100\n",
            "51/51 - 0s - loss: 0.0227 - accuracy: 0.9941 - val_loss: 0.0373 - val_accuracy: 0.9862\n",
            "Epoch 79/100\n",
            "51/51 - 0s - loss: 0.0228 - accuracy: 0.9941 - val_loss: 0.0369 - val_accuracy: 0.9862\n",
            "Epoch 80/100\n",
            "51/51 - 0s - loss: 0.0225 - accuracy: 0.9936 - val_loss: 0.0373 - val_accuracy: 0.9862\n",
            "Epoch 81/100\n",
            "51/51 - 0s - loss: 0.0225 - accuracy: 0.9941 - val_loss: 0.0368 - val_accuracy: 0.9881\n",
            "Epoch 82/100\n",
            "51/51 - 0s - loss: 0.0225 - accuracy: 0.9941 - val_loss: 0.0368 - val_accuracy: 0.9862\n",
            "Epoch 83/100\n",
            "51/51 - 0s - loss: 0.0222 - accuracy: 0.9941 - val_loss: 0.0363 - val_accuracy: 0.9881\n",
            "Epoch 84/100\n",
            "51/51 - 0s - loss: 0.0224 - accuracy: 0.9931 - val_loss: 0.0377 - val_accuracy: 0.9862\n",
            "Epoch 85/100\n",
            "51/51 - 0s - loss: 0.0222 - accuracy: 0.9941 - val_loss: 0.0378 - val_accuracy: 0.9862\n",
            "Epoch 86/100\n",
            "51/51 - 0s - loss: 0.0223 - accuracy: 0.9936 - val_loss: 0.0378 - val_accuracy: 0.9862\n",
            "Epoch 87/100\n",
            "51/51 - 0s - loss: 0.0220 - accuracy: 0.9941 - val_loss: 0.0379 - val_accuracy: 0.9862\n",
            "Epoch 88/100\n",
            "51/51 - 0s - loss: 0.0219 - accuracy: 0.9936 - val_loss: 0.0378 - val_accuracy: 0.9862\n",
            "Epoch 89/100\n",
            "51/51 - 0s - loss: 0.0218 - accuracy: 0.9946 - val_loss: 0.0374 - val_accuracy: 0.9862\n",
            "Epoch 90/100\n",
            "51/51 - 0s - loss: 0.0215 - accuracy: 0.9941 - val_loss: 0.0373 - val_accuracy: 0.9881\n",
            "Epoch 91/100\n",
            "51/51 - 0s - loss: 0.0218 - accuracy: 0.9936 - val_loss: 0.0387 - val_accuracy: 0.9881\n",
            "Epoch 92/100\n",
            "51/51 - 0s - loss: 0.0216 - accuracy: 0.9946 - val_loss: 0.0380 - val_accuracy: 0.9862\n",
            "Epoch 93/100\n",
            "51/51 - 0s - loss: 0.0213 - accuracy: 0.9941 - val_loss: 0.0376 - val_accuracy: 0.9862\n",
            "Epoch 94/100\n",
            "51/51 - 0s - loss: 0.0213 - accuracy: 0.9946 - val_loss: 0.0378 - val_accuracy: 0.9862\n",
            "Epoch 95/100\n",
            "51/51 - 0s - loss: 0.0222 - accuracy: 0.9941 - val_loss: 0.0386 - val_accuracy: 0.9862\n",
            "Epoch 96/100\n",
            "51/51 - 0s - loss: 0.0211 - accuracy: 0.9941 - val_loss: 0.0374 - val_accuracy: 0.9881\n",
            "Epoch 97/100\n",
            "51/51 - 0s - loss: 0.0212 - accuracy: 0.9941 - val_loss: 0.0377 - val_accuracy: 0.9862\n",
            "Epoch 98/100\n",
            "51/51 - 0s - loss: 0.0210 - accuracy: 0.9946 - val_loss: 0.0375 - val_accuracy: 0.9862\n",
            "Epoch 99/100\n",
            "51/51 - 0s - loss: 0.0210 - accuracy: 0.9946 - val_loss: 0.0375 - val_accuracy: 0.9881\n",
            "Epoch 100/100\n",
            "51/51 - 0s - loss: 0.0207 - accuracy: 0.9946 - val_loss: 0.0383 - val_accuracy: 0.9862\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "mOao2g3qZcxa"
      },
      "source": [
        "**Accuracy**"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "vHYlEUZ9ZWTR",
        "outputId": "91e6fede-a972-450a-ec93-1e58433fdb9a"
      },
      "source": [
        "# Get the loss and accuracy of the model by evaluation\n",
        "loss, acc = model.evaluate(X_test, y_test)\n",
        "\n",
        "# Print the loss and accuracy score for the model\n",
        "print(\"%s: %.2f%%\" % (model.metrics_names[0], loss*100))\n",
        "print(\"%s: %.2f%%\" % (model.metrics_names[1], acc*100))"
      ],
      "execution_count": 50,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "20/20 [==============================] - 0s 977us/step - loss: 0.0643 - accuracy: 0.9826\n",
            "loss: 6.43%\n",
            "accuracy: 98.26%\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "l6AQTXgtZhld",
        "outputId": "c792dd2d-dced-444c-e85b-d903e24ec77f"
      },
      "source": [
        "# Predicting the output predictions\n",
        "y_pred = model.predict(X_test).round()\n",
        "# Calculating the F1 score, recall, and precision\n",
        "print(\"%s: %.2f%%\" % (\"F1-score\", f1_score(y_test, y_pred)*100))\n",
        "print(\"%s: %.2f%%\" % (\"Recall\", recall_score(y_test, y_pred)*100))\n",
        "print(\"%s: %.2f%%\" % (\"Precision\", precision_score(y_test, y_pred)*100))"
      ],
      "execution_count": 51,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "F1-score: 84.06%\n",
            "Recall: 78.38%\n",
            "Precision: 90.62%\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "_9p5BgaoZuKm"
      },
      "source": [
        "**Training and validation loss**"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 295
        },
        "id": "lkWCnMFbZoOB",
        "outputId": "926adbfd-d79b-48ba-96d0-3f5c3ea6d88e"
      },
      "source": [
        "# summarize the result and plot the training and test loss\n",
        "plt.plot(result.history['loss'])\n",
        "plt.plot(result.history['val_loss'])\n",
        "\n",
        "# Set the parameters\n",
        "plt.title('Deep learning model loss')\n",
        "plt.ylabel('Loss')\n",
        "plt.xlabel('Epochs')\n",
        "plt.legend(['train', 'test'], loc='upper right')\n",
        "\n",
        "# Display the plots\n",
        "plt.show()\n"
      ],
      "execution_count": 52,
      "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": "z_wbpz0RnLen"
      },
      "source": [
        "**Training and validation accuracy**"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 295
        },
        "id": "WB8WKaURnQWc",
        "outputId": "dbea6c25-9625-493b-e3db-ce275f641418"
      },
      "source": [
        "# summarize the result and plot the training and test accuracy\n",
        "plt.plot(result.history['accuracy'])\n",
        "plt.plot(result.history['val_accuracy'])\n",
        "\n",
        "# Set the parameters\n",
        "plt.title('Deep learning model accuracy')\n",
        "plt.ylabel('Accuracy')\n",
        "plt.xlabel('Epochs')\n",
        "plt.legend(['train', 'test'], loc='upper right')\n",
        "\n",
        "# Display the plots\n",
        "plt.show()"
      ],
      "execution_count": 53,
      "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": "ucIQzOnLaBLH"
      },
      "source": [
        "**Conclusion**\n",
        "\n",
        "In this project, we have illustrated the potential use of neural networks in diagnosing hypothyroid disease. The identification of thyroid disease identiﬁcation remains to be an essential yet difﬁcult task in both clinical diagnosis and statistical classification. The diagnosis involves the use of a large number of interrelated patient attributes along with extremely unbalanced groups that result in a complicated relationship between the input features. However, Artiﬁcial neural networks show a promising result due to its flexible nature in modeling complex patterns of data for diagnosis. This knowledge can be further extended for many other disease diagnosis where such problems exist."
      ]
    }
  ]
}