water_quality_data_collection.ipynb

{
  "cells": [
    {
      "attachments": {},
      "cell_type": "markdown",
      "metadata": {
        "id": "eav8Z_I6Txhe"
      },
      "source": [
        "# Import Data\n"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 1,
      "metadata": {},
      "outputs": [
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "Requirement already satisfied: numpy in /opt/homebrew/lib/python3.11/site-packages (1.25.2)\n",
            "\n",
            "\u001b[1m[\u001b[0m\u001b[34;49mnotice\u001b[0m\u001b[1;39;49m]\u001b[0m\u001b[39;49m A new release of pip is available: \u001b[0m\u001b[31;49m23.1.2\u001b[0m\u001b[39;49m -> \u001b[0m\u001b[32;49m23.2.1\u001b[0m\n",
            "\u001b[1m[\u001b[0m\u001b[34;49mnotice\u001b[0m\u001b[1;39;49m]\u001b[0m\u001b[39;49m To update, run: \u001b[0m\u001b[32;49mpython3.11 -m pip install --upgrade pip\u001b[0m\n",
            "Note: you may need to restart the kernel to use updated packages.\n",
            "Requirement already satisfied: pandas in /opt/homebrew/lib/python3.11/site-packages (2.0.0)\n",
            "Requirement already satisfied: python-dateutil>=2.8.2 in /Users/kittizz/Library/Python/3.11/lib/python/site-packages (from pandas) (2.8.2)\n",
            "Requirement already satisfied: pytz>=2020.1 in /opt/homebrew/lib/python3.11/site-packages (from pandas) (2023.3)\n",
            "Requirement already satisfied: tzdata>=2022.1 in /opt/homebrew/lib/python3.11/site-packages (from pandas) (2023.3)\n",
            "Requirement already satisfied: numpy>=1.21.0 in /opt/homebrew/lib/python3.11/site-packages (from pandas) (1.25.2)\n",
            "Requirement already satisfied: six>=1.5 in /opt/homebrew/lib/python3.11/site-packages (from python-dateutil>=2.8.2->pandas) (1.16.0)\n",
            "\n",
            "\u001b[1m[\u001b[0m\u001b[34;49mnotice\u001b[0m\u001b[1;39;49m]\u001b[0m\u001b[39;49m A new release of pip is available: \u001b[0m\u001b[31;49m23.1.2\u001b[0m\u001b[39;49m -> \u001b[0m\u001b[32;49m23.2.1\u001b[0m\n",
            "\u001b[1m[\u001b[0m\u001b[34;49mnotice\u001b[0m\u001b[1;39;49m]\u001b[0m\u001b[39;49m To update, run: \u001b[0m\u001b[32;49mpython3.11 -m pip install --upgrade pip\u001b[0m\n",
            "Note: you may need to restart the kernel to use updated packages.\n",
            "Requirement already satisfied: matplotlib in /opt/homebrew/lib/python3.11/site-packages (3.7.1)\n",
            "Requirement already satisfied: contourpy>=1.0.1 in /opt/homebrew/lib/python3.11/site-packages (from matplotlib) (1.0.7)\n",
            "Requirement already satisfied: cycler>=0.10 in /opt/homebrew/lib/python3.11/site-packages (from matplotlib) (0.11.0)\n",
            "Requirement already satisfied: fonttools>=4.22.0 in /opt/homebrew/lib/python3.11/site-packages (from matplotlib) (4.39.3)\n",
            "Requirement already satisfied: kiwisolver>=1.0.1 in /opt/homebrew/lib/python3.11/site-packages (from matplotlib) (1.4.4)\n",
            "Requirement already satisfied: numpy>=1.20 in /opt/homebrew/lib/python3.11/site-packages (from matplotlib) (1.25.2)\n",
            "Requirement already satisfied: packaging>=20.0 in /Users/kittizz/Library/Python/3.11/lib/python/site-packages (from matplotlib) (23.0)\n",
            "Requirement already satisfied: pillow>=6.2.0 in /opt/homebrew/lib/python3.11/site-packages (from matplotlib) (10.0.0)\n",
            "Requirement already satisfied: pyparsing>=2.3.1 in /opt/homebrew/lib/python3.11/site-packages (from matplotlib) (3.0.9)\n",
            "Requirement already satisfied: python-dateutil>=2.7 in /Users/kittizz/Library/Python/3.11/lib/python/site-packages (from matplotlib) (2.8.2)\n",
            "Requirement already satisfied: six>=1.5 in /opt/homebrew/lib/python3.11/site-packages (from python-dateutil>=2.7->matplotlib) (1.16.0)\n",
            "\n",
            "\u001b[1m[\u001b[0m\u001b[34;49mnotice\u001b[0m\u001b[1;39;49m]\u001b[0m\u001b[39;49m A new release of pip is available: \u001b[0m\u001b[31;49m23.1.2\u001b[0m\u001b[39;49m -> \u001b[0m\u001b[32;49m23.2.1\u001b[0m\n",
            "\u001b[1m[\u001b[0m\u001b[34;49mnotice\u001b[0m\u001b[1;39;49m]\u001b[0m\u001b[39;49m To update, run: \u001b[0m\u001b[32;49mpython3.11 -m pip install --upgrade pip\u001b[0m\n",
            "Note: you may need to restart the kernel to use updated packages.\n",
            "Requirement already satisfied: xlrd in /opt/homebrew/lib/python3.11/site-packages (2.0.1)\n",
            "\n",
            "\u001b[1m[\u001b[0m\u001b[34;49mnotice\u001b[0m\u001b[1;39;49m]\u001b[0m\u001b[39;49m A new release of pip is available: \u001b[0m\u001b[31;49m23.1.2\u001b[0m\u001b[39;49m -> \u001b[0m\u001b[32;49m23.2.1\u001b[0m\n",
            "\u001b[1m[\u001b[0m\u001b[34;49mnotice\u001b[0m\u001b[1;39;49m]\u001b[0m\u001b[39;49m To update, run: \u001b[0m\u001b[32;49mpython3.11 -m pip install --upgrade pip\u001b[0m\n",
            "Note: you may need to restart the kernel to use updated packages.\n",
            "Requirement already satisfied: scikit-learn in /opt/homebrew/lib/python3.11/site-packages (1.2.2)\n",
            "Requirement already satisfied: numpy>=1.17.3 in /opt/homebrew/lib/python3.11/site-packages (from scikit-learn) (1.25.2)\n",
            "Requirement already satisfied: scipy>=1.3.2 in /opt/homebrew/lib/python3.11/site-packages (from scikit-learn) (1.10.1)\n",
            "Requirement already satisfied: joblib>=1.1.1 in /opt/homebrew/lib/python3.11/site-packages (from scikit-learn) (1.2.0)\n",
            "Requirement already satisfied: threadpoolctl>=2.0.0 in /opt/homebrew/lib/python3.11/site-packages (from scikit-learn) (3.1.0)\n",
            "\n",
            "\u001b[1m[\u001b[0m\u001b[34;49mnotice\u001b[0m\u001b[1;39;49m]\u001b[0m\u001b[39;49m A new release of pip is available: \u001b[0m\u001b[31;49m23.1.2\u001b[0m\u001b[39;49m -> \u001b[0m\u001b[32;49m23.2.1\u001b[0m\n",
            "\u001b[1m[\u001b[0m\u001b[34;49mnotice\u001b[0m\u001b[1;39;49m]\u001b[0m\u001b[39;49m To update, run: \u001b[0m\u001b[32;49mpython3.11 -m pip install --upgrade pip\u001b[0m\n",
            "Note: you may need to restart the kernel to use updated packages.\n",
            "Requirement already satisfied: seaborn in /opt/homebrew/lib/python3.11/site-packages (0.12.2)\n",
            "Requirement already satisfied: numpy!=1.24.0,>=1.17 in /opt/homebrew/lib/python3.11/site-packages (from seaborn) (1.25.2)\n",
            "Requirement already satisfied: pandas>=0.25 in /opt/homebrew/lib/python3.11/site-packages (from seaborn) (2.0.0)\n",
            "Requirement already satisfied: matplotlib!=3.6.1,>=3.1 in /opt/homebrew/lib/python3.11/site-packages (from seaborn) (3.7.1)\n",
            "Requirement already satisfied: contourpy>=1.0.1 in /opt/homebrew/lib/python3.11/site-packages (from matplotlib!=3.6.1,>=3.1->seaborn) (1.0.7)\n",
            "Requirement already satisfied: cycler>=0.10 in /opt/homebrew/lib/python3.11/site-packages (from matplotlib!=3.6.1,>=3.1->seaborn) (0.11.0)\n",
            "Requirement already satisfied: fonttools>=4.22.0 in /opt/homebrew/lib/python3.11/site-packages (from matplotlib!=3.6.1,>=3.1->seaborn) (4.39.3)\n",
            "Requirement already satisfied: kiwisolver>=1.0.1 in /opt/homebrew/lib/python3.11/site-packages (from matplotlib!=3.6.1,>=3.1->seaborn) (1.4.4)\n",
            "Requirement already satisfied: packaging>=20.0 in /Users/kittizz/Library/Python/3.11/lib/python/site-packages (from matplotlib!=3.6.1,>=3.1->seaborn) (23.0)\n",
            "Requirement already satisfied: pillow>=6.2.0 in /opt/homebrew/lib/python3.11/site-packages (from matplotlib!=3.6.1,>=3.1->seaborn) (10.0.0)\n",
            "Requirement already satisfied: pyparsing>=2.3.1 in /opt/homebrew/lib/python3.11/site-packages (from matplotlib!=3.6.1,>=3.1->seaborn) (3.0.9)\n",
            "Requirement already satisfied: python-dateutil>=2.7 in /Users/kittizz/Library/Python/3.11/lib/python/site-packages (from matplotlib!=3.6.1,>=3.1->seaborn) (2.8.2)\n",
            "Requirement already satisfied: pytz>=2020.1 in /opt/homebrew/lib/python3.11/site-packages (from pandas>=0.25->seaborn) (2023.3)\n",
            "Requirement already satisfied: tzdata>=2022.1 in /opt/homebrew/lib/python3.11/site-packages (from pandas>=0.25->seaborn) (2023.3)\n",
            "Requirement already satisfied: six>=1.5 in /opt/homebrew/lib/python3.11/site-packages (from python-dateutil>=2.7->matplotlib!=3.6.1,>=3.1->seaborn) (1.16.0)\n",
            "\n",
            "\u001b[1m[\u001b[0m\u001b[34;49mnotice\u001b[0m\u001b[1;39;49m]\u001b[0m\u001b[39;49m A new release of pip is available: \u001b[0m\u001b[31;49m23.1.2\u001b[0m\u001b[39;49m -> \u001b[0m\u001b[32;49m23.2.1\u001b[0m\n",
            "\u001b[1m[\u001b[0m\u001b[34;49mnotice\u001b[0m\u001b[1;39;49m]\u001b[0m\u001b[39;49m To update, run: \u001b[0m\u001b[32;49mpython3.11 -m pip install --upgrade pip\u001b[0m\n",
            "Note: you may need to restart the kernel to use updated packages.\n"
          ]
        }
      ],
      "source": [
        "%pip install numpy\n",
        "%pip install pandas\n",
        "%pip install matplotlib\n",
        "%pip install xlrd\n",
        "%pip install scikit-learn\n",
        "%pip install seaborn\n"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 3,
      "metadata": {
        "id": "3TEPV63DTwOO"
      },
      "outputs": [],
      "source": [
        "import numpy as np\n",
        "import pandas as pd\n",
        "import matplotlib.pyplot as plt\n",
        "from pathlib import Path\n",
        "from sklearn.model_selection import train_test_split\n",
        "import seaborn as sns  # สร้างกราฟ Boxplot\n",
        "from scipy.stats import zscore\n",
        "\n",
        "%matplotlib inline\n"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 194,
      "metadata": {},
      "outputs": [],
      "source": [
        "# Define the directory to search for CSV and XLS files\n",
        "directory = Path(r'./dataset/')\n",
        "# Find all CSV and XLS files in the directory and its subdirectories\n",
        "csv = list(directory.rglob('*.csv'))\n",
        "xls = list(directory.rglob('*.xls'))\n",
        "\n",
        "files = csv + xls\n"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 196,
      "metadata": {},
      "outputs": [],
      "source": [
        "# import urllib.parse\n",
        "# dataset_repo = \"https://github.com/kittizz/water_quality_data_collection/raw/main/\"\n",
        "csv_files = list()\n",
        "xls_files = list()\n",
        "for index in range(len(files)):\n",
        "    file = files[index]\n",
        "    if \"63\" in file.name:\n",
        "        # if file.name.endswith(\".csv\"):\n",
        "        #     csv_files.append(dataset_repo+urllib.parse.quote(str(file)))\n",
        "        # if file.name.endswith(\".xls\"):\n",
        "        #     xls_files.append(dataset_repo+urllib.parse.quote(str(file)))\n",
        "\n",
        "        if file.name.endswith(\".csv\"):\n",
        "            csv_files.append(str(file))\n",
        "        if file.name.endswith(\".xls\"):\n",
        "            xls_files.append(str(file))\n"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 199,
      "metadata": {
        "id": "sxLcLFDElkSa"
      },
      "outputs": [],
      "source": [
        "df1 = pd.concat([pd.read_csv(f, encoding='utf-8')\n",
        "                 for f in csv_files])\n",
        "df2 = pd.concat([pd.read_excel(f)\n",
        "                 for f in xls_files])\n",
        "\n",
        "df_concat = pd.concat([df1, df2], axis=0)\n"
      ]
    },
    {
      "attachments": {},
      "cell_type": "markdown",
      "metadata": {
        "id": "HoBWH9Tj0GME"
      },
      "source": [
        "# Preprocessing - Data Cleaning\n"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 200,
      "metadata": {},
      "outputs": [
        {
          "data": {
            "text/plain": [
              "(418131, 11)"
            ]
          },
          "execution_count": 200,
          "metadata": {},
          "output_type": "execute_result"
        }
      ],
      "source": [
        "df_concat.shape"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 201,
      "metadata": {},
      "outputs": [
        {
          "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>สถานี</th>\n",
              "      <th>วันที่-เวลา</th>\n",
              "      <th>pH</th>\n",
              "      <th>DO</th>\n",
              "      <th>EC</th>\n",
              "      <th>Temp</th>\n",
              "      <th>หมายเลข</th>\n",
              "      <th>Salinity</th>\n",
              "      <th>Turbidity</th>\n",
              "      <th>BOD</th>\n",
              "      <th>COD</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>0</th>\n",
              "      <td>นครชัยศรี จ.นครปฐม (121)</td>\n",
              "      <td>1 ม.ค.2563 00:00</td>\n",
              "      <td>-</td>\n",
              "      <td>2.300</td>\n",
              "      <td>464.700</td>\n",
              "      <td>28.3</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1</th>\n",
              "      <td>นครชัยศรี จ.นครปฐม (121)</td>\n",
              "      <td>1 ม.ค.2563 00:30</td>\n",
              "      <td>-</td>\n",
              "      <td>2.300</td>\n",
              "      <td>463.100</td>\n",
              "      <td>28.3</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2</th>\n",
              "      <td>นครชัยศรี จ.นครปฐม (121)</td>\n",
              "      <td>1 ม.ค.2563 01:00</td>\n",
              "      <td>-</td>\n",
              "      <td>2.300</td>\n",
              "      <td>464.200</td>\n",
              "      <td>28.3</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>3</th>\n",
              "      <td>นครชัยศรี จ.นครปฐม (121)</td>\n",
              "      <td>1 ม.ค.2563 01:30</td>\n",
              "      <td>-</td>\n",
              "      <td>2.300</td>\n",
              "      <td>466.700</td>\n",
              "      <td>28.3</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>4</th>\n",
              "      <td>นครชัยศรี จ.นครปฐม (121)</td>\n",
              "      <td>1 ม.ค.2563 02:00</td>\n",
              "      <td>-</td>\n",
              "      <td>2.300</td>\n",
              "      <td>468.200</td>\n",
              "      <td>28.2</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>...</th>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>10215</th>\n",
              "      <td>NaN</td>\n",
              "      <td>31 ก.ค.2563 21:30</td>\n",
              "      <td>8.1</td>\n",
              "      <td>4.9</td>\n",
              "      <td>238.2</td>\n",
              "      <td>30.7</td>\n",
              "      <td>222.0</td>\n",
              "      <td>0.1</td>\n",
              "      <td></td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>10216</th>\n",
              "      <td>NaN</td>\n",
              "      <td>31 ก.ค.2563 22:00</td>\n",
              "      <td>8.1</td>\n",
              "      <td>4.9</td>\n",
              "      <td>237.4</td>\n",
              "      <td>30.7</td>\n",
              "      <td>222.0</td>\n",
              "      <td>0.1</td>\n",
              "      <td></td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>10217</th>\n",
              "      <td>NaN</td>\n",
              "      <td>31 ก.ค.2563 22:30</td>\n",
              "      <td>8.1</td>\n",
              "      <td>4.9</td>\n",
              "      <td>236.7</td>\n",
              "      <td>30.7</td>\n",
              "      <td>222.0</td>\n",
              "      <td>0.1</td>\n",
              "      <td></td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>10218</th>\n",
              "      <td>NaN</td>\n",
              "      <td>31 ก.ค.2563 23:00</td>\n",
              "      <td>8.1</td>\n",
              "      <td>4.9</td>\n",
              "      <td>237.5</td>\n",
              "      <td>30.7</td>\n",
              "      <td>222.0</td>\n",
              "      <td>0.1</td>\n",
              "      <td></td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>10219</th>\n",
              "      <td>NaN</td>\n",
              "      <td>31 ก.ค.2563 23:30</td>\n",
              "      <td>8.1</td>\n",
              "      <td>4.9</td>\n",
              "      <td>237.4</td>\n",
              "      <td>30.7</td>\n",
              "      <td>222.0</td>\n",
              "      <td>0.1</td>\n",
              "      <td></td>\n",
              "      <td>NaN</td>\n",
              "      <td>NaN</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "<p>418131 rows × 11 columns</p>\n",
              "</div>"
            ],
            "text/plain": [
              "                           สถานี         วันที่-เวลา   pH     DO       EC   \n",
              "0        นครชัยศรี จ.นครปฐม (121)   1 ม.ค.2563 00:00    -  2.300  464.700  \\\n",
              "1        นครชัยศรี จ.นครปฐม (121)   1 ม.ค.2563 00:30    -  2.300  463.100   \n",
              "2        นครชัยศรี จ.นครปฐม (121)   1 ม.ค.2563 01:00    -  2.300  464.200   \n",
              "3        นครชัยศรี จ.นครปฐม (121)   1 ม.ค.2563 01:30    -  2.300  466.700   \n",
              "4        นครชัยศรี จ.นครปฐม (121)   1 ม.ค.2563 02:00    -  2.300  468.200   \n",
              "...                           ...                ...  ...    ...      ...   \n",
              "10215                         NaN  31 ก.ค.2563 21:30  8.1    4.9    238.2   \n",
              "10216                         NaN  31 ก.ค.2563 22:00  8.1    4.9    237.4   \n",
              "10217                         NaN  31 ก.ค.2563 22:30  8.1    4.9    236.7   \n",
              "10218                         NaN  31 ก.ค.2563 23:00  8.1    4.9    237.5   \n",
              "10219                         NaN  31 ก.ค.2563 23:30  8.1    4.9    237.4   \n",
              "\n",
              "       Temp  หมายเลข Salinity Turbidity  BOD  COD  \n",
              "0      28.3      NaN      NaN       NaN  NaN  NaN  \n",
              "1      28.3      NaN      NaN       NaN  NaN  NaN  \n",
              "2      28.3      NaN      NaN       NaN  NaN  NaN  \n",
              "3      28.3      NaN      NaN       NaN  NaN  NaN  \n",
              "4      28.2      NaN      NaN       NaN  NaN  NaN  \n",
              "...     ...      ...      ...       ...  ...  ...  \n",
              "10215  30.7    222.0      0.1            NaN  NaN  \n",
              "10216  30.7    222.0      0.1            NaN  NaN  \n",
              "10217  30.7    222.0      0.1            NaN  NaN  \n",
              "10218  30.7    222.0      0.1            NaN  NaN  \n",
              "10219  30.7    222.0      0.1            NaN  NaN  \n",
              "\n",
              "[418131 rows x 11 columns]"
            ]
          },
          "execution_count": 201,
          "metadata": {},
          "output_type": "execute_result"
        }
      ],
      "source": [
        "df_concat"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 202,
      "metadata": {},
      "outputs": [],
      "source": [
        "df = df_concat.copy()"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 203,
      "metadata": {},
      "outputs": [
        {
          "data": {
            "text/plain": [
              "Index(['สถานี ', 'วันที่-เวลา', 'pH', 'DO', 'EC', 'Temp', 'หมายเลข',\n",
              "       'Salinity', 'Turbidity', 'BOD', 'COD'],\n",
              "      dtype='object')"
            ]
          },
          "execution_count": 203,
          "metadata": {},
          "output_type": "execute_result"
        }
      ],
      "source": [
        "df.columns"
      ]
    },
    {
      "attachments": {},
      "cell_type": "markdown",
      "metadata": {
        "id": "LXMIo2RE0qAt"
      },
      "source": [
        "## missing value\n"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 204,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "Hkenryba0gdD",
        "outputId": "076f426c-2035-485f-b7da-1df1190a2925"
      },
      "outputs": [
        {
          "data": {
            "text/plain": [
              "สถานี          352596\n",
              "วันที่-เวลา         0\n",
              "pH                  0\n",
              "DO                  0\n",
              "EC                  0\n",
              "Temp                0\n",
              "หมายเลข         65535\n",
              "Salinity       181163\n",
              "Turbidity      249931\n",
              "BOD            341389\n",
              "COD            341389\n",
              "dtype: int64"
            ]
          },
          "execution_count": 204,
          "metadata": {},
          "output_type": "execute_result"
        }
      ],
      "source": [
        "# ตรวจสอบ missing value\n",
        "df.isnull().sum()"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 205,
      "metadata": {},
      "outputs": [],
      "source": [
        "# ลบ column ที่เป็นภาษาไทย และ วันที่ทิ้งไป\n",
        "df = df[['pH', 'DO', 'EC', 'Temp',\n",
        "         'Salinity', 'Turbidity', 'BOD', 'COD']]"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 206,
      "metadata": {},
      "outputs": [
        {
          "data": {
            "text/plain": [
              "418131"
            ]
          },
          "execution_count": 206,
          "metadata": {},
          "output_type": "execute_result"
        }
      ],
      "source": [
        "# นับจำนวนข้อมูลที่ไม่เป็น missing value ของ column  pH, DO, EC, Temp\n",
        "df[['pH', 'DO', 'EC', 'Temp']].notnull().all(axis=1).sum()"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 207,
      "metadata": {},
      "outputs": [],
      "source": [
        "# ลบ column Salinity, Turbidity, BOD, COD ทิ้งไป เพราะว่ามีข้อมูลไม่ครบจากทุก dataset\n",
        "df.drop(columns=['Salinity', 'Turbidity',\n",
        "        'BOD', 'COD'], inplace=True)"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 208,
      "metadata": {},
      "outputs": [
        {
          "data": {
            "text/plain": [
              "pH      0\n",
              "DO      0\n",
              "EC      0\n",
              "Temp    0\n",
              "dtype: int64"
            ]
          },
          "execution_count": 208,
          "metadata": {},
          "output_type": "execute_result"
        }
      ],
      "source": [
        "# ตรวจสอบจำนวน missing value\n",
        "df.isna().sum()"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 209,
      "metadata": {},
      "outputs": [],
      "source": [
        "# ลบข้อมูลที่มี missing value ออกไป\n",
        "df.dropna(inplace=True)"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 210,
      "metadata": {},
      "outputs": [
        {
          "data": {
            "text/plain": [
              "pH      0\n",
              "DO      0\n",
              "EC      0\n",
              "Temp    0\n",
              "dtype: int64"
            ]
          },
          "execution_count": 210,
          "metadata": {},
          "output_type": "execute_result"
        }
      ],
      "source": [
        "# ตรวจสอบจำนวน missing value\n",
        "df.isnull().sum()"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 211,
      "metadata": {},
      "outputs": [],
      "source": [
        "# แปลงตัวเลขที่เป็น string ให้เป็นตัวเลข และ เปลี่ยนตัวอักษรที่ไม่ถูกต้องให้เป็น NaN\n",
        "for col in df.columns:\n",
        "    df[col] = pd.to_numeric(df[col], errors='coerce')"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 212,
      "metadata": {},
      "outputs": [],
      "source": [
        "# ลบข้อมูลที่มี missing value ออกไป\n",
        "df.dropna(inplace=True)"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 213,
      "metadata": {},
      "outputs": [
        {
          "data": {
            "text/plain": [
              "pH      0\n",
              "DO      0\n",
              "EC      0\n",
              "Temp    0\n",
              "dtype: int64"
            ]
          },
          "execution_count": 213,
          "metadata": {},
          "output_type": "execute_result"
        }
      ],
      "source": [
        "df.isnull().sum()"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 214,
      "metadata": {},
      "outputs": [],
      "source": [
        "# เปลี่ยนชนิดข้อมูลให้เป็น float\n",
        "df = df.astype(float)\n",
        "\n",
        "# เปลี่ยนชนิดข้อมูลให้ pH เป็น int\n",
        "df['pH'] = df['pH'].astype(int)"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 215,
      "metadata": {},
      "outputs": [],
      "source": [
        "\n",
        "# df = df.iloc[:].apply(lambda x: x.round(2))"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 216,
      "metadata": {},
      "outputs": [],
      "source": [
        "# reset index\n",
        "df = df.reindex()"
      ]
    },
    {
      "attachments": {},
      "cell_type": "markdown",
      "metadata": {
        "id": "LaGv8c0s07U5"
      },
      "source": [
        "## Outlier data\n"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 217,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 300
        },
        "id": "AObeEgAt1jmz",
        "outputId": "feb327b3-aee5-4a99-84ca-a84bcf7d6bbe"
      },
      "outputs": [
        {
          "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>pH</th>\n",
              "      <th>DO</th>\n",
              "      <th>EC</th>\n",
              "      <th>Temp</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>count</th>\n",
              "      <td>319952.000000</td>\n",
              "      <td>319952.000000</td>\n",
              "      <td>319952.000000</td>\n",
              "      <td>319952.000000</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>mean</th>\n",
              "      <td>5.823480</td>\n",
              "      <td>4.425241</td>\n",
              "      <td>1050.342042</td>\n",
              "      <td>29.592894</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>std</th>\n",
              "      <td>2.205455</td>\n",
              "      <td>2.438073</td>\n",
              "      <td>7180.436664</td>\n",
              "      <td>3.602120</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>min</th>\n",
              "      <td>-2.000000</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>-105.000000</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>25%</th>\n",
              "      <td>5.000000</td>\n",
              "      <td>2.400000</td>\n",
              "      <td>230.300000</td>\n",
              "      <td>28.048200</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>50%</th>\n",
              "      <td>6.000000</td>\n",
              "      <td>4.500000</td>\n",
              "      <td>316.500000</td>\n",
              "      <td>29.968500</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>75%</th>\n",
              "      <td>7.000000</td>\n",
              "      <td>6.400000</td>\n",
              "      <td>473.690000</td>\n",
              "      <td>31.600000</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>max</th>\n",
              "      <td>14.000000</td>\n",
              "      <td>19.350000</td>\n",
              "      <td>100000.000000</td>\n",
              "      <td>200.000000</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>"
            ],
            "text/plain": [
              "                  pH             DO             EC           Temp\n",
              "count  319952.000000  319952.000000  319952.000000  319952.000000\n",
              "mean        5.823480       4.425241    1050.342042      29.592894\n",
              "std         2.205455       2.438073    7180.436664       3.602120\n",
              "min        -2.000000       0.000000       0.000000    -105.000000\n",
              "25%         5.000000       2.400000     230.300000      28.048200\n",
              "50%         6.000000       4.500000     316.500000      29.968500\n",
              "75%         7.000000       6.400000     473.690000      31.600000\n",
              "max        14.000000      19.350000  100000.000000     200.000000"
            ]
          },
          "execution_count": 217,
          "metadata": {},
          "output_type": "execute_result"
        }
      ],
      "source": [
        "# แสดงค่าสถิติพื้นฐาน\n",
        "df.describe()"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 218,
      "metadata": {},
      "outputs": [
        {
          "data": {
            "text/plain": [
              "<Axes: >"
            ]
          },
          "execution_count": 218,
          "metadata": {},
          "output_type": "execute_result"
        },
        {
          "data": {
            "image/png": "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",
            "text/plain": [
              "<Figure size 640x480 with 1 Axes>"
            ]
          },
          "metadata": {},
          "output_type": "display_data"
        }
      ],
      "source": [
        "df.boxplot()"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 219,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 283
        },
        "id": "USD-HJGM1U-I",
        "outputId": "b574e4fc-9724-4be6-9204-fc09c6aef404"
      },
      "outputs": [
        {
          "data": {
            "image/png": "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",
            "text/plain": [
              "<Figure size 640x480 with 1 Axes>"
            ]
          },
          "metadata": {},
          "output_type": "display_data"
        }
      ],
      "source": [
        "# แสดง box plot เพื่อตรวจสอบ Outlier Data\n",
        "plt.boxplot(df['pH'], labels=['pH'])\n",
        "plt.show()"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 220,
      "metadata": {},
      "outputs": [
        {
          "data": {
            "image/png": "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",
            "text/plain": [
              "<Figure size 640x480 with 1 Axes>"
            ]
          },
          "metadata": {},
          "output_type": "display_data"
        }
      ],
      "source": [
        "plt.boxplot(df['DO'], labels=['DO'])\n",
        "plt.show()"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 221,
      "metadata": {},
      "outputs": [
        {
          "data": {
            "image/png": "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",
            "text/plain": [
              "<Figure size 640x480 with 1 Axes>"
            ]
          },
          "metadata": {},
          "output_type": "display_data"
        }
      ],
      "source": [
        "plt.boxplot(df['EC'], labels=['EC'])\n",
        "plt.show()"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 222,
      "metadata": {},
      "outputs": [
        {
          "data": {
            "image/png": "iVBORw0KGgoAAAANSUhEUgAAAjMAAAGdCAYAAADnrPLBAAAAOXRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjcuMSwgaHR0cHM6Ly9tYXRwbG90bGliLm9yZy/bCgiHAAAACXBIWXMAAA9hAAAPYQGoP6dpAAAqa0lEQVR4nO3df3DU9Z3H8dfuhoSEZDcFya8hSDAov4IieDTUOKRkDEg9cyHXE5ABiyBI2oNQlHAK4lXCIKhtlaJeLU4H/FEmxjFX6XEomhsi1ViswYIBEonND6hANuFHQrJ7f3SydTFoxM1+88k+HzPfMfv9vHf3HWdgX3y+n+9nbV6v1ysAAABD2a1uAAAA4NsgzAAAAKMRZgAAgNEIMwAAwGiEGQAAYDTCDAAAMBphBgAAGI0wAwAAjBZmdQPB4PF4VFdXp5iYGNlsNqvbAQAA3eD1etXc3KykpCTZ7ZeffwmJMFNXV6fk5GSr2wAAAFegtrZWQ4YMuex4SISZmJgYSX//n+F0Oi3uBgAAdIfb7VZycrLvc/xyQiLMdF5acjqdhBkAAAzzdUtEWAAMAACMRpgBAABGI8wAAACjEWYAAIDRCDMAAMBohBkAAGA0wgwAADAaYQYAABgtJDbNA9A3dXR0qKysTPX19UpMTFRGRoYcDofVbQEIsh6dmSkqKtJNN92kmJgYxcXFKScnR4cPH/aruXDhgpYuXapBgwYpOjpaM2fOVGNjo1/N8ePHNWPGDEVFRSkuLk4rV65Ue3t7T7YOoJcrLi5WamqqMjMzNXv2bGVmZio1NVXFxcVWtwYgyHo0zLz99ttaunSp3n33Xe3evVsXL17UrbfeqrNnz/pqli9frtdff12/+93v9Pbbb6uurk65ubm+8Y6ODs2YMUNtbW3at2+fXnjhBW3btk1r1qzpydYB9GLFxcXKy8tTWlqaysvL1dzcrPLycqWlpSkvL49AA4QYm9fr9QbrzU6ePKm4uDi9/fbbuuWWW9TU1KTBgwdrx44dysvLkyQdOnRIo0aNUnl5ub773e/qjTfe0A9+8APV1dUpPj5ekrR161Y98MADOnnypMLDw7/2fd1ut1wul5qamvhuJsBwHR0dSk1NVVpamkpKSmS3/+PfZB6PRzk5OaqsrFRVVRWXnADDdffzO6gLgJuamiRJAwcOlCRVVFTo4sWLysrK8tWMHDlSQ4cOVXl5uST5/rXVGWQkKTs7W263WwcPHuzyfVpbW+V2u/0OAH1DWVmZampqtHr1ar8gI0l2u12FhYWqrq5WWVmZRR0CCLaghRmPx6Nly5bpe9/7nsaOHStJamhoUHh4uGJjY/1q4+Pj1dDQ4Kv5YpDpHO8c60pRUZFcLpfvSE5ODvBvA8Aq9fX1kuT7e+RSnec76wD0fUELM0uXLlVlZaVeeumlHn+vwsJCNTU1+Y7a2toef08AwZGYmChJqqys7HK883xnHYC+LyhhJj8/X6WlpXrrrbc0ZMgQ3/mEhAS1tbXpzJkzfvWNjY1KSEjw1Vx6d1Pn486aS0VERMjpdPodAPqGjIwMDRs2TOvXr5fH4/Eb83g8KioqUkpKijIyMizqEECw9WiY8Xq9ys/P16uvvqo333xTKSkpfuMTJkxQv379tGfPHt+5w4cP6/jx40pPT5ckpaen66OPPtKJEyd8Nbt375bT6dTo0aN7sn0AvZDD4dDmzZtVWlqqnJwcv7uZcnJyVFpaqk2bNrH4FwghPXo303333acdO3botdde03XXXec773K5FBkZKUlasmSJfv/732vbtm1yOp368Y9/LEnat2+fpL/fuXDDDTcoKSlJGzduVENDg+bOnat77rlH69ev71Yf3M0E9D3FxcVasWKFampqfOdSUlK0adMmv+0dAJiru5/fPRpmbDZbl+d/85vfaP78+ZL+vmneihUr9OKLL6q1tVXZ2dnasmWL3yWkTz/9VEuWLNHevXs1YMAAzZs3Txs2bFBYWPc2MCbMAH0TOwADfVuvCDO9BWEGAADz9Mp9ZgAAAAKNMAMAAIxGmAEAAEYjzAAAAKMRZgAAgNEIMwAAwGjd26gFAHoh9pkBIDEzA8BQxcXFSk1NVWZmpmbPnq3MzEylpqaquLjY6tYABBlhBoBxiouLlZeXp7S0NL/vZkpLS1NeXh6BBggx7AAMwCgdHR1KTU1VWlqaSkpKZLf/499kHo9HOTk5qqysVFVVFZecAMOxAzCAPqmsrEw1NTVavXq1X5CRJLvdrsLCQlVXV6usrMyiDgEEG2EGgFHq6+slSWPHju1yvPN8Zx2Avo8wA8AoiYmJkqTKysouxzvPd9YB6PsIMwCMkpGRoWHDhmn9+vXyeDx+Yx6PR0VFRUpJSVFGRoZFHQIINsIMAKM4HA5t3rxZpaWlysnJ8bubKScnR6Wlpdq0aROLf4EQwqZ5AIyTm5urnTt3asWKFZo8ebLvfEpKinbu3Knc3FwLuwMQbNyaDcBY7AAM9G3d/fxmZgaAsRwOh6ZMmWJ1GwAsxpoZAABgNMIMAAAwGmEGAAAYjTADAACMRpgBAABGI8wAAACjEWYAAIDRCDMAAMBohBkAAGA0wgwAADAaYQYAABiNMAMAAIxGmAEAAEYjzAAAAKMRZgAAgNEIMwAAwGiEGQAAYDTCDAAAMBphBgAAGI0wAwAAjEaYAQAARiPMAAAAoxFmAACA0QgzAADAaIQZAABgNMIMAAAwWo+GmXfeeUe33367kpKSZLPZVFJS4jc+f/582Ww2v2PatGl+NadOndKcOXPkdDoVGxurBQsWqKWlpSfbBgAABunRMHP27Fldf/31evrppy9bM23aNNXX1/uOF1980W98zpw5OnjwoHbv3q3S0lK98847WrRoUU+2DQAADBLWky8+ffp0TZ8+/StrIiIilJCQ0OXYX/7yF+3atUvvvfeeJk6cKEn65S9/qdtuu02bNm1SUlJSwHsGAABmsXzNzN69exUXF6frrrtOS5Ys0eeff+4bKy8vV2xsrC/ISFJWVpbsdrv2799vRbsAAKCX6dGZma8zbdo05ebmKiUlRUePHtXq1as1ffp0lZeXy+FwqKGhQXFxcX7PCQsL08CBA9XQ0HDZ121tbVVra6vvsdvt7rHfAQAAWMvSMHPnnXf6fk5LS9O4ceN0zTXXaO/evZo6deoVv25RUZHWrVsXiBYBAEAvZ/llpi8aPny4rrrqKh05ckSSlJCQoBMnTvjVtLe369SpU5ddZyNJhYWFampq8h21tbU92jcAALCOpTMzl/rss8/0+eefKzExUZKUnp6uM2fOqKKiQhMmTJAkvfnmm/J4PJo0adJlXyciIkIRERFB6RmAdTo6OlRWVqb6+nolJiYqIyNDDofD6rYABFmPhpmWlhbfLIskVVdX68CBAxo4cKAGDhyodevWaebMmUpISNDRo0d1//33KzU1VdnZ2ZKkUaNGadq0aVq4cKG2bt2qixcvKj8/X3feeSd3MgEhrri4WCtWrFBNTY3v3LBhw7R582bl5uZa1xiAoOvRy0zvv/++xo8fr/Hjx0uSCgoKNH78eK1Zs0YOh0N//vOf9c///M+69tprtWDBAk2YMEFlZWV+syrbt2/XyJEjNXXqVN122226+eab9eyzz/Zk2wB6ueLiYuXl5SktLU3l5eVqbm5WeXm50tLSlJeXp+LiYqtbBBBENq/X67W6iZ7mdrvlcrnU1NQkp9NpdTsAvoWOjg6lpqYqLS1NJSUlstv/8W8yj8ejnJwcVVZWqqqqiktOgOG6+/ndqxYAA8DXKSsrU01NjVavXu0XZCTJbrersLBQ1dXVKisrs6hDAMFGmAFglPr6eknS2LFjuxzvPN9ZB6DvI8wAMErn3Y6VlZVdjnee76wD0PcRZgAYJSMjQ8OGDdP69evl8Xj8xjwej4qKipSSkqKMjAyLOgQQbIQZAEZxOBzavHmzSktLlZOT43c3U05OjkpLS7Vp0yYW/wIhpFdtmgcA3ZGbm6udO3dqxYoVmjx5su98SkqKdu7cyT4zQIjh1mwAxmIHYKBv6+7nNzMzAIzlcDg0ZcoUq9sAYDHWzAAAAKMRZgAAgNEIMwAAwGismQFgLBYAA5CYmQFgqOLiYqWmpiozM1OzZ89WZmamUlNT+cZsIAQRZgAYp7i4WHl5eUpLS/PbNC8tLU15eXkEGiDEsM8MAKN0dHQoNTVVaWlpKikp8fvmbI/Ho5ycHFVWVqqqqopLToDhuvv5zcwMAKOUlZWppqZGq1ev9gsykmS321VYWKjq6mqVlZVZ1CGAYCPMADBKfX29JGns2LFdjnee76wD0PcRZgAYJTExUZJUWVnZ5Xjn+c46AH0fYQaAUTIyMjRs2DCtX79eHo/Hb8zj8aioqEgpKSnKyMiwqEMAwUaYAWAUh8OhzZs3q7S0VDk5OX53M+Xk5Ki0tFSbNm1i8S8QQtg0D4BxcnNztXPnTq1YsUKTJ0/2nU9JSdHOnTuVm5trYXcAgo2ZGQDGunRniUsvOwEIDYQZAMbp3DRv3LhxfpeZxo0bx6Z5QAhi0zwARmHTPCB0sGkegD6JTfMAXIowA8AobJoH4FKEGQBGYdM8AJcizAAwCpvmAbgUYQaAUdg0D8Cl2DQPgHHYNA/AF3FrNgBjtbS0aO7cuTp69KiuueYa/fa3v1V0dLTVbQEIkO5+fjMzA8BIOTk5eu2113yPP/roI8XExOiOO+5QSUmJdY0BCDrWzAAwTmeQCQ8P16pVq3TkyBGtWrVK4eHheu2115STk2N1iwCCiMtMAIxy/vx5RUVFKTw8XM3NzQoPD/eNtbW1KSYmRm1tbTp37pwiIyMt7BTAt8UOwAD6pJUrV0qSCgoK/IKMJIWHh2vZsmV+dQD6PsIMAKNUVVVJku65554uxxcsWOBXB6DvI8wAMMqIESMkSf/1X/+ljo4O7d27Vy+++KL27t2rjo4O/frXv/arA9D3sWYGgFE618yEhYUpKSlJx48f940NHTpUdXV1am9vZ80M0AewZgZAnxQZGambbrpJ7e3tqq2t1V133aUPPvhAd911l2pra9Xe3q6bbrqJIAOEEGZmABilo6NDqampOn/+vBobG780Hh8fr6ioKFVVVfGVBoDhmJkB0CeVlZWppqZGJSUlOnfunJYuXapbb71VS5cu1blz5/Tqq6+qurpaZWVlVrcKIEjYARiAUerr6yVJY8eOlcPhUGpqqmw2m6655ho5HA6NHTvWrw5A30eYAWCUxMRESdLixYv18ssvq7293Te2cuVK/fCHP/SrA9D3cZkJgFEyMjLkdDq1fft2DRo0SM8995zq6+v13HPPadCgQdqxY4ecTqcyMjKsbhVAkPRomHnnnXd0++23KykpSTab7Utf/ub1erVmzRolJiYqMjJSWVlZX9ro6tSpU5ozZ46cTqdiY2O1YMECtbS09GTbAHqxjo4O398BEydO1JgxYzRgwACNGTNGEydOlPT3b9Pu6Oiwsk0AQdSjYebs2bO6/vrr9fTTT3c5vnHjRv3iF7/Q1q1btX//fg0YMEDZ2dm6cOGCr2bOnDk6ePCgdu/erdLSUr3zzjtatGhRT7YNoBfbsmWLPB6PlixZooMHD2ry5MlyOp2aPHmyPv74Yy1evFgej0dbtmyxulUAweINEkneV1991ffY4/F4ExISvI899pjv3JkzZ7wRERHeF1980ev1er0ff/yxV5L3vffe89W88cYbXpvN5v3rX//a7fduamrySvI2NTV9+18EgKXy8/O9krz19fXe9vZ271tvveXdsWOH96233vK2t7d7//rXv3olefPz861uFcC31N3Pb8vWzFRXV6uhoUFZWVm+cy6XS5MmTVJ5ebkkqby8XLGxsb6pY0nKysqS3W7X/v37g94zAOtdc801kqTS0lI5HA5NmTJFs2bN0pQpU+RwOFRaWupXB6DvsyzMNDQ0SPr7BldfFB8f7xtraGhQXFyc33hYWJgGDhzoq+lKa2ur3G633wGgb7jvvvsUFhamBx980O9OJklqb2/XmjVrFBYWpvvuu8+iDgEEW5+8m6moqEgul8t3JCcnW90SgAAJDw/X8uXL1djYqCFDhujZZ59VXV2dnn32WQ0ZMkSNjY1avny5wsPDrW4VQJBYts9MQkKCJKmxsdFvP4jGxkbdcMMNvpoTJ074Pa+9vV2nTp3yPb8rhYWFKigo8D12u90EGqAP2bhxoyTpiSee0L333us7HxYWppUrV/rGAYQGy2ZmUlJSlJCQoD179vjOud1u7d+/X+np6ZKk9PR0nTlzRhUVFb6aN998Ux6PR5MmTbrsa0dERMjpdPodAPqWjRs36uzZs3riiSeUn5+vJ554QmfPniXIACGoR2dmWlpadOTIEd/j6upqHThwQAMHDtTQoUO1bNky/exnP9OIESOUkpKihx56SElJScrJyZEkjRo1StOmTdPChQu1detWXbx4Ufn5+brzzjuVlJTUk60DMEB4eLiWLVtmdRsALNajYeb9999XZmam73HnpZ958+Zp27Ztuv/++3X27FktWrRIZ86c0c0336xdu3apf//+vuds375d+fn5mjp1qux2u2bOnKlf/OIXPdk2AAAwiM3r9XqtbqKndfcrxAEAQO/R3c/vPnk3EwAACB2EGQAAYDTCDAAAMBphBgAAGI0wAwAAjEaYAQAARiPMAAAAoxFmAACA0QgzAADAaIQZAABgNMIMAAAwGmEGAAAYjTADAACMRpgBAABGI8wAAACjEWYAAIDRwqxuAACuVEdHh8rKylRfX6/ExERlZGTI4XBY3RaAIGNmBoCRiouLlZqaqszMTM2ePVuZmZlKTU1VcXGx1a0BCDLCDADjFBcXKy8vT2lpaSovL1dzc7PKy8uVlpamvLw8Ag0QYmxer9drdRM9ze12y+VyqampSU6n0+p2AHwLHR0dSk1NVVpamkpKSmS3/+PfZB6PRzk5OaqsrFRVVRWXnADDdffzm5kZAEYpKytTTU2NVq9e7RdkJMlut6uwsFDV1dUqKyuzqEMAwcYCYABGqa+vlySNHTu2ywXAY8eO9asD0PcRZgAYJTExUZL01FNP6ZlnnlFNTY1vbNiwYVq0aJFfHYC+j8tMAIySkZGhuLg4FRYWauzYsX4LgMeOHavVq1crLi5OGRkZVrcKIEgIMwCM88X7Frxer+8AEJoIMwCMUlZWppMnT6qoqEiVlZWaPHmynE6nJk+erIMHD2r9+vU6ceIEC4CBEEKYAWCUzoW9ycnJX5qN8Xg8Gjp0qF8dgL6PBcAAjNK5sHfu3Lm67bbblJOTo/PnzysyMlJHjhzR3Llz/eoA9H1smgfAKG1tbRowYIDCw8PV1tam9vZ231hYWJjv/NmzZxUeHm5hpwC+LTbNA9An7du3T+3t7Tp37pzsdrtWrVqlqqoqrVq1Sna7XefOnVN7e7v27dtndasAgoTLTACMUltbK0lyOp2KjY3Vhg0btGHDBknS1VdfrdOnT8vtdvvqAPR9zMwAMMr+/fslSVOnTv3SmNfr1fe//32/OgB9HzMzAIzSuczv1VdfVWRkpN/YyZMnVVJS4lcHoO8jzAAwyvDhw30/x8TE6L777tPw4cN17Ngx/fa3v9X58+e/VAegbyPMADDK6NGjJf39G7JPnDihzZs3+43b7XZ5PB5fHYC+jzUzAIzSeZeSx+PpcrzzPHczAaGDMAPAKJcLMVdaB8B8hBkARomNjQ1oHQDzEWYAGOXUqVMBrQNgPsIMAKPU1NQEtA6A+QgzAIzyySefBLQOgPkIMwCM0tzcHNA6AOYjzAAwSnR0dEDrAJiPMAPAKKmpqQGtA2A+wgwAo/ztb38LaB0A81keZh5++GHZbDa/Y+TIkb7xCxcuaOnSpRo0aJCio6M1c+ZMNTY2WtgxACt1988/f08AocPyMCNJY8aMUX19ve/4v//7P9/Y8uXL9frrr+t3v/ud3n77bdXV1Sk3N9fCbgFYqbvfhs23ZgOho1d80WRYWJgSEhK+dL6pqUm//vWvtWPHDn3/+9+XJP3mN7/RqFGj9O677+q73/1usFsFYLF+/foFtA6A+XrFzExVVZWSkpI0fPhwzZkzR8ePH5ckVVRU6OLFi8rKyvLVjhw5UkOHDlV5efllX6+1tVVut9vvANA3dPfPM3/ugdBheZiZNGmStm3bpl27dulXv/qVqqurlZGRoebmZjU0NCg8PPxL37ESHx+vhoaGy75mUVGRXC6X70hOTu7h3wJAsDAzA+BSll9mmj59uu/ncePGadKkSbr66qv1yiuvKDIy8opes7CwUAUFBb7HbrebQAP0EYMGDVJVVVW36gCEBstnZi4VGxura6+9VkeOHFFCQoLa2tp05swZv5rGxsYu19h0ioiIkNPp9DsA9A1XXXVVQOsAmK/XhZmWlhYdPXpUiYmJmjBhgvr166c9e/b4xg8fPqzjx48rPT3dwi4BWKWlpSWgdQDMZ/llpp/+9Ke6/fbbdfXVV6uurk5r166Vw+HQrFmz5HK5tGDBAhUUFGjgwIFyOp368Y9/rPT0dO5kAkLURx99FNA6AOazPMx89tlnmjVrlj7//HMNHjxYN998s959910NHjxYkvTEE0/Ibrdr5syZam1tVXZ2trZs2WJx1wCs8vnnnwe0DoD5bN4Q2FnK7XbL5XKpqamJ9TOA4Ww2W7drQ+CvN6BP6+7nd69bMwMAAPBNEGYAAIDRCDMAAMBohBkAAGA0wgwAADAaYQYAABjN8n1mAISec+fO6dChQz3+Ph988ME3fs7IkSMVFRXVA90A6CmEGQBBd+jQIU2YMKHH3+dK3qOiokI33nhjD3QDoKcQZgAE3ciRI1VRUXFFz/0mAeVK3mPkyJHf+DkArEWYARB0UVFRQZn9YIYFCA0sAAZglO5+RQFfZQCEDsIMAON8XVAhyAChhctMAL6RqqoqNTc3W92GKioqulw/U1FRcUV3MQVaTEyMRowYYXUbQEggzADotqqqKl177bVWt/GVgnGXVHd98sknBBogCAgzALqt5fRJjU+w62c/+5lSUlKsbkeS1Nraqrq6OiUlJSkiIsLqdiRJ1dXVevDBB9Vy+qQkwgzQ0wgzALqtf8txfXBvtFS7Qaq1upt/uEHqVf2MknTbvdH6S8txSZOtbgfo8wgzALrttOMq3fhMix566KFesx9Lb56Z+fVtQ61uBQgJhBkA3fZxVY3+1OBR7tJ1VrdihOjvDLa6BSAkEGYAdFtOTo6k3vP9RZe7m6k34G4mIHhs3hDYkMHtdsvlcqmpqUlOp9PqdgAEgM1mu+xYCPy1BoSE7n5+s2keAON8VZDpzjiAvoUwA8Ao3Q0qBBogdLBmBkDQnTt3TocOHerx97mSnYB7y3ogAN1HmAEQdIcOHQrKTr1X8h4VFRV82zZgGMIMgKAbOXLkFd919E0CypW8R2/ZPwdA9xFmAARdVFRUUGY/mGEBQgMLgAEAgNEIMwAAwGiEGQAAYDTCDAAAMBphBgAAGI0wAwAAjEaYAQAARiPMAAAAoxFmAACA0QgzAADAaIQZAABgNMIMAAAwGmEGAAAYjTADwChXX311QOsAmI8wAwAAjEaYAWCUTz/9NKB1AMxHmAEAAEYjzAAAAKMZE2aefvppDRs2TP3799ekSZP0xz/+0eqWAABAL2BEmHn55ZdVUFCgtWvX6oMPPtD111+v7OxsnThxwurWAACAxYwIM48//rgWLlyou+++W6NHj9bWrVsVFRWl559/3urWAACAxXp9mGlra1NFRYWysrJ85+x2u7KyslReXt7lc1pbW+V2u/0OAADQN/X6MPO3v/1NHR0dio+P9zsfHx+vhoaGLp9TVFQkl8vlO5KTk4PRKgAAsECvDzNXorCwUE1NTb6jtrbW6pYAAEAPCbO6ga9z1VVXyeFwqLGx0e98Y2OjEhISunxORESEIiIigtEeAACwWK+fmQkPD9eECRO0Z88e3zmPx6M9e/YoPT3dws4AWGHYsGEBrQNgvl4/MyNJBQUFmjdvniZOnKh/+qd/0pNPPqmzZ8/q7rvvtro1AEHW3t4e0DoA5jMizPzbv/2bTp48qTVr1qihoUE33HCDdu3a9aVFwQD6vsGDB+uzzz7rVh2A0NDrLzN1ys/P16effqrW1lbt379fkyZNsrolABYYMmRIQOsAmM+YMAMAkjR8+PCA1gEwH2EGgFFef/31gNYBMB9hBoBRurtvFPtLAaGDMAPAKBcvXgxoHQDzEWYAGMVmswW0DoD5CDMAjBIdHR3QOgDmI8wAMEp2dnZA6wCYjzADwCj/+7//G9A6AOYjzAAwSktLS0DrAJiPMAPAKB0dHb6fL/3Kgri4uC7rAPRthBkAxjp58qTf4xMnTljUCQArEWYAGCUiIiKgdQDMR5gBYJTMzMyA1gEwH2EGgFHuvffegNYBMB9hBoBRXnnllYDWATAfYQaAUY4dOxbQOgDmI8wAMEp371jiziYgdBBmABjl9OnTAa0DYD7CDACjnD9/PqB1AMxHmAFgFLu9e39tdbcOgPn40w7AKF6vN6B1AMxHmAFgFMIMgEsRZgAYJTo6OqB1AMxHmAFglH/5l38JaB0A89m8ITAX63a75XK51NTUJKfTaXU7AL6F8+fPKyoq6mvrzp07p8jIyCB0BKCndPfzm5kZAEaJjIzUHXfc8ZU1d9xxB0EGCCGEGQDGKSkpuWygueOOO1RSUhLchgBYKszqBgDgSpSUlOj8+fNauXKlqqqqNGLECD322GPMyAAhiDADwFiRkZF66qmnrG4DgMW4zAQAAIxGmAEAAEYjzAAAAKMRZgAAgNEIMwAAwGiEGQAAYDTCDAAAMBphBgAAGI0wAwAAjEaYAQAARiPMAAAAoxFmAACA0QgzAADAaIQZAABgNMIMAAAwmqVhZtiwYbLZbH7Hhg0b/Gr+/Oc/KyMjQ/3791dycrI2btxoUbcAAKA3CrO6gUceeUQLFy70PY6JifH97Ha7deuttyorK0tbt27VRx99pB/96EeKjY3VokWLrGgXAAD0MpaHmZiYGCUkJHQ5tn37drW1ten5559XeHi4xowZowMHDujxxx8nzAAAAEm9YM3Mhg0bNGjQII0fP16PPfaY2tvbfWPl5eW65ZZbFB4e7juXnZ2tw4cP6/Tp05d9zdbWVrndbr8DAAD0TZbOzPzkJz/RjTfeqIEDB2rfvn0qLCxUfX29Hn/8cUlSQ0ODUlJS/J4THx/vG/vOd77T5esWFRVp3bp1Pds8AADoFQI+M7Nq1aovLeq99Dh06JAkqaCgQFOmTNG4ceO0ePFibd68Wb/85S/V2tr6rXooLCxUU1OT76itrQ3ErwYAAHqhgM/MrFixQvPnz//KmuHDh3d5ftKkSWpvb1dNTY2uu+46JSQkqLGx0a+m8/Hl1tlIUkREhCIiIr5Z4wAAwEgBDzODBw/W4MGDr+i5Bw4ckN1uV1xcnCQpPT1d//Ef/6GLFy+qX79+kqTdu3fruuuuu+wlJgAAEFosWwBcXl6uJ598Uh9++KGOHTum7du3a/ny5brrrrt8QWX27NkKDw/XggULdPDgQb388sv6+c9/roKCAqvaBgAAvYxlC4AjIiL00ksv6eGHH1Zra6tSUlK0fPlyv6Dicrn0P//zP1q6dKkmTJigq666SmvWrOG2bAAA4GPzer1eq5voaW63Wy6XS01NTXI6nVa3AwAAuqG7n9+W7zMDAADwbRBmAACA0QgzAADAaIQZAABgNMIMAAAwGmEGAAAYjTADAACMRpgBAABGI8wAAACjEWYAAIDRCDMAAMBohBkAAGA0wgwAADAaYQYAABiNMAMAAIxGmAEAAEYjzAAAAKMRZgAAgNEIMwAAwGiEGQAAYDTCDAAAMBphBgAAGI0wAwAAjEaYAQAARiPMAAAAoxFmAACA0QgzAADAaIQZAABgNMIMAAAwGmEGAAAYjTADAACMRpgBAABGI8wAAACjEWYAAIDRCDMAAMBohBkAAGA0wgwAADAaYQYAABiNMAMAAIxGmAEAAEYjzAAAAKMRZgAAgNEIMwAAwGiEGQAAYLQeCzOPPvqoJk+erKioKMXGxnZZc/z4cc2YMUNRUVGKi4vTypUr1d7e7lezd+9e3XjjjYqIiFBqaqq2bdvWUy0DMEx1dbUiIyNlt9sVGRmp6upqq1sCYIEeCzNtbW3613/9Vy1ZsqTL8Y6ODs2YMUNtbW3at2+fXnjhBW3btk1r1qzx1VRXV2vGjBnKzMzUgQMHtGzZMt1zzz36wx/+0FNtAzCEw+HQ8OHDdeHCBXm9Xl24cEHDhw+Xw+GwujUAQWbzer3ennyDbdu2admyZTpz5ozf+TfeeEM/+MEPVFdXp/j4eEnS1q1b9cADD+jkyZMKDw/XAw88oP/+7/9WZWWl73l33nmnzpw5o127dnW7B7fbLZfLpaamJjmdzoD8XgCs43A45PF4Ljtut9vV0dERxI4A9ITufn5btmamvLxcaWlpviAjSdnZ2XK73Tp48KCvJisry+952dnZKi8v/8rXbm1tldvt9jsA9A3V1dVfGWQkyePxcMkJCCGWhZmGhga/ICPJ97ihoeEra9xut86fP3/Z1y4qKpLL5fIdycnJAe4egFWGDx8e0DoA5vtGYWbVqlWy2WxfeRw6dKineu22wsJCNTU1+Y7a2lqrWwIAAD0k7JsUr1ixQvPnz//Kmu7+ayghIUF//OMf/c41Njb6xjr/23nuizVOp1ORkZGXfe2IiAhFRER0qw8AZuvXr59WrVqlDRs26OLFi1a3A8AC3yjMDB48WIMHDw7IG6enp+vRRx/ViRMnFBcXJ0navXu3nE6nRo8e7av5/e9/7/e83bt3Kz09PSA9ADBbRUWFbrzxRknSI488og8++EATJkywuCsAwdZja2aOHz+uAwcO6Pjx4+ro6NCBAwd04MABtbS0SJJuvfVWjR49WnPnztWHH36oP/zhD3rwwQe1dOlS36zK4sWLdezYMd1///06dOiQtmzZoldeeUXLly/vqbYBGGTChAlyOp3avHmznE4nQQYIUT12a/b8+fP1wgsvfOn8W2+9pSlTpkiSPv30Uy1ZskR79+7VgAEDNG/ePG3YsEFhYf+YMNq7d6+WL1+ujz/+WEOGDNFDDz30tZe6LsWt2UDfYbPZul3bwztPAOhh3f387vF9ZnoDwgzQd9jt9m6FFJvN9rW3cAPo3Xr9PjMAcCW+uIlmIOoAmI8wA8AonTcIBKoOgPkIMwCM83WXmULg6jmALyDMADCS1+vVwYMHZbf//a8xu92ugwcPEmSAEPSN9pkBgN5k9OjRfKEkAGZmAACA2QgzAADAaIQZAABgNMIMAAAwGmEGAAAYjTADAACMRpgBAABGI8wAAACjEWYAAIDRQmIH4M7tzd1ut8WdAACA7ur83P66rykJiTDT3NwsSUpOTra4EwAA8E01NzfL5XJddtzmDYFvZfN4PKqrq1NMTIxsNpvV7QAIILfbreTkZNXW1srpdFrdDoAA8nq9am5uVlJSku9LZbsSEmEGQN/ldrvlcrnU1NREmAFCFAuAAQCA0QgzAADAaIQZAEaLiIjQ2rVrFRERYXUrACzCmhkAAGA0ZmYAAIDRCDMAAMBohBkAAGA0wgwAADAaYQaAJWw221ceDz/8sNUtAjBESHw3E4Dep76+3vfzyy+/rDVr1ujw4cO+c9HR0Va0BcBAzMwAsERCQoLvcLlcstlsfudeeukljRo1Sv3799fIkSO1ZcsW33Nrampks9n0yiuvKCMjQ5GRkbrpppv0ySef6L333tPEiRMVHR2t6dOn6+TJk77nzZ8/Xzk5OVq3bp0GDx4sp9OpxYsXq62tzYr/BQAChJkZAL3O9u3btWbNGj311FMaP368/vSnP2nhwoUaMGCA5s2b56tbu3atnnzySQ0dOlQ/+tGPNHv2bMXExOjnP/+5oqKi9MMf/lBr1qzRr371K99z9uzZo/79+2vv3r2qqanR3XffrUGDBunRRx+14lcFEACEGQC9ztq1a7V582bl5uZKklJSUvTxxx/rmWee8QszP/3pT5WdnS1J+vd//3fNmjVLe/bs0fe+9z1J0oIFC7Rt2za/1w4PD9fzzz+vqKgojRkzRo888ohWrlyp//zP//zKb+UF0HsRZgD0KmfPntXRo0e1YMECLVy40He+vb1dLpfLr3bcuHG+n+Pj4yVJaWlpfudOnDjh95zrr79eUVFRvsfp6elqaWlRbW2trr766oD+LgCCgzADoFdpaWmRJD333HOaNGmS35jD4fB73K9fP9/PNputy3Mej6enWgXQSxBmAPQq8fHxSkpK0rFjxzRnzpyAv/6HH36o8+fPKzIyUpL07rvvKjo6WsnJyQF/LwDBQZgB0OusW7dOP/nJT+RyuTRt2jS1trbq/fff1+nTp1VQUPCtXrutrU0LFizQgw8+qJqaGq1du1b5+fmslwEMRpgB0Ovcc889ioqK0mOPPaaVK1dqwIABSktL07Jly771a0+dOlUjRozQLbfcotbWVs2aNYsN+gDD2bxer9fqJgAgGObPn68zZ86opKTE6lYABBDzqgAAwGiEGQAAYDQuMwEAAKMxMwMAAIxGmAEAAEYjzAAAAKMRZgAAgNEIMwAAwGiEGQAAYDTCDAAAMBphBgAAGI0wAwAAjPb/r5dTvfk+tQMAAAAASUVORK5CYII=",
            "text/plain": [
              "<Figure size 640x480 with 1 Axes>"
            ]
          },
          "metadata": {},
          "output_type": "display_data"
        }
      ],
      "source": [
        "plt.boxplot(df['Temp'], labels=['Temp'])\n",
        "plt.show()"
      ]
    },
    {
      "attachments": {},
      "cell_type": "markdown",
      "metadata": {
        "id": "MvbgOZdP7H9X"
      },
      "source": [
        "จัดการ outlier data\n"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 223,
      "metadata": {
        "id": "aLW2u4iU5n66"
      },
      "outputs": [],
      "source": [
        "def remove_outlier(df_in, col_name):\n",
        "    q1 = np.quantile(df_in[col_name], 0.25)\n",
        "    q3 = np.quantile(df_in[col_name], 0.75)\n",
        "    iqr = q3-q1  # Interquartile range\n",
        "    fence_low = q1-1.5*iqr\n",
        "    fence_high = q3+1.5*iqr\n",
        "    print('Lower bound: ', fence_low)\n",
        "    print('Upper bound: ', fence_high)\n",
        "    df_out = df_in.loc[(df_in[col_name] > fence_low) &\n",
        "                       (df_in[col_name] < fence_high)]\n",
        "    return df_out"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 224,
      "metadata": {},
      "outputs": [
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "Lower bound:  2.0\n",
            "Upper bound:  10.0\n"
          ]
        },
        {
          "data": {
            "image/png": "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",
            "text/plain": [
              "<Figure size 640x480 with 1 Axes>"
            ]
          },
          "metadata": {},
          "output_type": "display_data"
        }
      ],
      "source": [
        "df_out = df.copy()\n",
        "df_out = remove_outlier(df_out, 'pH')\n",
        "\n",
        "plt.boxplot(df_out['pH'])\n",
        "plt.show()"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 225,
      "metadata": {},
      "outputs": [
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "Lower bound:  -3.5561625000000006\n",
            "Upper bound:  12.3269375\n",
            "Lower bound:  -3.492600000000001\n",
            "Upper bound:  12.221\n"
          ]
        },
        {
          "data": {
            "image/png": "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",
            "text/plain": [
              "<Figure size 640x480 with 1 Axes>"
            ]
          },
          "metadata": {},
          "output_type": "display_data"
        }
      ],
      "source": [
        "\n",
        "df_out = remove_outlier(df_out, 'DO')\n",
        "df_out = remove_outlier(df_out, 'DO')\n",
        "plt.boxplot(df_out['DO'])\n",
        "plt.show()"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 226,
      "metadata": {},
      "outputs": [
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "Lower bound:  -100.25000000000006\n",
            "Upper bound:  796.95\n",
            "Lower bound:  -63.97\n",
            "Upper bound:  721.582\n",
            "Lower bound:  -58.30000000000001\n",
            "Upper bound:  709.7\n",
            "Lower bound:  -56.49999999999997\n",
            "Upper bound:  705.9\n",
            "Lower bound:  -56.107500000000016\n",
            "Upper bound:  704.8645\n",
            "Lower bound:  -55.950000000000045\n",
            "Upper bound:  704.45\n",
            "Lower bound:  -55.950000000000045\n",
            "Upper bound:  704.45\n"
          ]
        },
        {
          "data": {
            "image/png": "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",
            "text/plain": [
              "<Figure size 640x480 with 1 Axes>"
            ]
          },
          "metadata": {},
          "output_type": "display_data"
        }
      ],
      "source": [
        "\n",
        "df_out = remove_outlier(df_out, 'EC')\n",
        "df_out = remove_outlier(df_out, 'EC')\n",
        "df_out = remove_outlier(df_out, 'EC')\n",
        "df_out = remove_outlier(df_out, 'EC')\n",
        "df_out = remove_outlier(df_out, 'EC')\n",
        "df_out = remove_outlier(df_out, 'EC')\n",
        "df_out = remove_outlier(df_out, 'EC')\n",
        "# df_out = remove_outlier(df_out, 'EC')\n",
        "# df_out = remove_outlier(df_out, 'EC')\n",
        "plt.boxplot(df_out['EC'])\n",
        "plt.show()"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 227,
      "metadata": {},
      "outputs": [
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "Lower bound:  22.299999999999997\n",
            "Upper bound:  36.7\n",
            "Lower bound:  22.800187499999993\n",
            "Upper bound:  36.399887500000006\n"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "Lower bound:  23.049999999999997\n",
            "Upper bound:  36.25\n"
          ]
        },
        {
          "data": {
            "image/png": "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",
            "text/plain": [
              "<Figure size 640x480 with 1 Axes>"
            ]
          },
          "metadata": {},
          "output_type": "display_data"
        }
      ],
      "source": [
        "\n",
        "df_out = remove_outlier(df_out, 'Temp')\n",
        "df_out = remove_outlier(df_out, 'Temp')\n",
        "df_out = remove_outlier(df_out, 'Temp')\n",
        "# df_out = remove_outlier(df_out, 'Temp')\n",
        "# df_out = remove_outlier(df_out, 'Temp')\n",
        "plt.boxplot(df_out['Temp'])\n",
        "plt.show()"
      ]
    },
    {
      "attachments": {},
      "cell_type": "markdown",
      "metadata": {
        "id": "LsjeNPhQ7b1h"
      },
      "source": [
        "## duplicate data\n"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 228,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "2O3PQSQA7OQ0",
        "outputId": "20b91e56-bfd3-433d-9302-994768839ddb"
      },
      "outputs": [
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "จำนวน Duplicate Data df: 46043\n"
          ]
        }
      ],
      "source": [
        "print('จำนวน Duplicate Data df:', df.duplicated().sum())"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 229,
      "metadata": {
        "id": "b5A3CUAb78wd"
      },
      "outputs": [],
      "source": [
        "# ลบ Duplicate Data\n",
        "df = df.drop_duplicates()"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 230,
      "metadata": {},
      "outputs": [],
      "source": [
        "df_split_copy = df.copy()"
      ]
    },
    {
      "attachments": {},
      "cell_type": "markdown",
      "metadata": {},
      "source": [
        "## แบ่งชุดข้อมูล\n"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 231,
      "metadata": {},
      "outputs": [],
      "source": [
        "# seed สุ่มข้อมูล\n",
        "seed = 5712\n",
        "# แบ่งชุดข้อมูลออกเป็น test\n",
        "df_split, df_test = train_test_split(\n",
        "    df_split_copy, test_size=0.0001, random_state=seed)\n",
        "\n",
        "# บ่งชุดข้อมูลออกเป็น  validation\n",
        "df_split, df_valid = train_test_split(\n",
        "    df_split, test_size=0.0001, random_state=seed)\n",
        "\n",
        "# บ่งชุดข้อมูลออกเป็น  training\n",
        "df_split, df_train = train_test_split(\n",
        "    df_split, test_size=0.01, random_state=seed)"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 232,
      "metadata": {},
      "outputs": [
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "จำนวนชุดข้อมูลทั้งหมด : 273909\n",
            "จำนวนชุดข้อมูล training : 2739\n",
            "จำนวนชุดข้อมูล validation : 28\n",
            "จำนวนชุดข้อมูล test : 28\n"
          ]
        }
      ],
      "source": [
        "# จำนวนชุดข้อมูล\n",
        "print('จำนวนชุดข้อมูลทั้งหมด :', len(df))\n",
        "print('จำนวนชุดข้อมูล training :', len(df_train))\n",
        "print('จำนวนชุดข้อมูล validation :', len(df_valid))\n",
        "print('จำนวนชุดข้อมูล test :', len(df_test))"
      ]
    },
    {
      "attachments": {},
      "cell_type": "markdown",
      "metadata": {
        "id": "Jzq1yaJsAGy_"
      },
      "source": [
        "## Data Transformation: Normalization\n"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 233,
      "metadata": {
        "id": "ihha_tsM_87e"
      },
      "outputs": [],
      "source": [
        "def minmaxNorm(Data, _min, _max):\n",
        "    Data_Norm = (Data - _min)/(_max - _min)\n",
        "    return Data_Norm\n",
        "\n",
        "\n",
        "def min4norm(Data):\n",
        "    _min = Data.min(axis=0, keepdims=True)\n",
        "    return _min\n",
        "\n",
        "\n",
        "def max4norm(Data):\n",
        "    _max = Data.max(axis=0, keepdims=True)\n",
        "    return _max"
      ]
    },
    {
      "attachments": {},
      "cell_type": "markdown",
      "metadata": {},
      "source": [
        "# KNN\n"
      ]
    },
    {
      "attachments": {},
      "cell_type": "markdown",
      "metadata": {},
      "source": [
        "### วัดระยะห่างระหว่างข้อมูลแต่ละตัวใน Validation Set กับข้อมูลทุกตัวใน Training Set\n"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 234,
      "metadata": {},
      "outputs": [],
      "source": [
        "def KNN_find_distance(X_Train, x_valid):\n",
        "    distance2 = ((X_Train - x_valid)**2).sum(axis=1)\n",
        "    distance = np.sqrt(distance2)\n",
        "    return distance"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 235,
      "metadata": {},
      "outputs": [],
      "source": [
        "def KNN_find_distance_each_data(X_Train, X_Valid):\n",
        "    all_distance = []\n",
        "    for x_valid in X_Valid:\n",
        "        distance = KNN_find_distance(X_Train, x_valid)\n",
        "        all_distance.append(distance)\n",
        "    return all_distance"
      ]
    },
    {
      "attachments": {},
      "cell_type": "markdown",
      "metadata": {},
      "source": [
        "### เรียงลำดับ Target ตามระยะห่าง\n"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 236,
      "metadata": {},
      "outputs": [],
      "source": [
        "def KNN_sort_target_by_distance(Y_Train, distance):\n",
        "    sorted_target = Y_Train[distance.argsort()]\n",
        "    return sorted_target"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 237,
      "metadata": {},
      "outputs": [],
      "source": [
        "def KNN_find_sorted_target_each_data(Y_Train, all_distance):\n",
        "    all_sorted_target = []\n",
        "    for distance in all_distance:\n",
        "        sorted_target = KNN_sort_target_by_distance(Y_Train, distance)\n",
        "        all_sorted_target.append(sorted_target)\n",
        "    return all_sorted_target"
      ]
    },
    {
      "attachments": {},
      "cell_type": "markdown",
      "metadata": {},
      "source": [
        "### หาผลลัพธ์จากเพื่อนบ้านที่ใกล้ที่สุด K ตัว\n"
      ]
    },
    {
      "attachments": {},
      "cell_type": "markdown",
      "metadata": {},
      "source": [
        "#### KNN-C\n"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 238,
      "metadata": {},
      "outputs": [],
      "source": [
        "def KNNC_find_class(sorted_target, K):\n",
        "    unique, count_unique = np.unique(sorted_target[:K, :], return_counts=True)\n",
        "    _class = unique[count_unique.argmax()]\n",
        "    return _class"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 239,
      "metadata": {},
      "outputs": [],
      "source": [
        "def KNNC_find_class_each_data(all_sorted_target, K):\n",
        "    all_class = []\n",
        "    for sorted_target in all_sorted_target:\n",
        "        _class = KNNC_find_class(sorted_target, K)\n",
        "        all_class.append(_class)\n",
        "    all_class = np.array(all_class).reshape(-1, 1)\n",
        "    return all_class"
      ]
    },
    {
      "attachments": {},
      "cell_type": "markdown",
      "metadata": {},
      "source": [
        "### หา error ของ K แต่ละค่า และ หา K ที่ทำให้ error บน Validation Set ต่ำที่สุด\n"
      ]
    },
    {
      "attachments": {},
      "cell_type": "markdown",
      "metadata": {},
      "source": [
        "#### KNN-C\n"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 240,
      "metadata": {},
      "outputs": [],
      "source": [
        "def find_error_classification(Y, Yhat):\n",
        "    N = Y.shape[0]\n",
        "    error = (100/N)*(Y != Yhat).sum()\n",
        "    return error"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 241,
      "metadata": {},
      "outputs": [],
      "source": [
        "def KNNC_find_error_each_K(Y_Valid, all_sorted_target, min_K, max_K):\n",
        "    K_list = [K for K in range(min_K, max_K + 1)]\n",
        "    error_list = []\n",
        "    for K in K_list:\n",
        "        K_all_class = KNNC_find_class_each_data(all_sorted_target, K)\n",
        "        K_error = find_error_classification(Y_Valid, K_all_class)\n",
        "        error_list.append(K_error)\n",
        "    return K_list, error_list"
      ]
    },
    {
      "attachments": {},
      "cell_type": "markdown",
      "metadata": {},
      "source": [
        "### เรียนรู้\n"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 242,
      "metadata": {},
      "outputs": [],
      "source": [
        "def KNN_find_best_K(K_list, error_list):\n",
        "    K_list = np.array(K_list)\n",
        "    error_list = np.array(error_list)\n",
        "    plt.plot(K_list, error_list)\n",
        "    plt.xlabel('K')\n",
        "    plt.ylabel('Error')\n",
        "    sorted_K = K_list[error_list.argsort()]\n",
        "    error_list.sort()\n",
        "    best_K = sorted_K[0]\n",
        "    L = len(K_list)\n",
        "    # for l in range(L):\n",
        "    # print('K :', sorted_K[l], ', error :', error_list[l])\n",
        "    return best_K"
      ]
    },
    {
      "attachments": {},
      "cell_type": "markdown",
      "metadata": {},
      "source": [
        "#### KNN-C\n"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 243,
      "metadata": {},
      "outputs": [],
      "source": [
        "def KNNC_fit(X_Train, Y_Train, X_Valid, Y_Valid, min_K, max_K):\n",
        "    _min = min4norm(X_Train)\n",
        "    _max = max4norm(X_Train)\n",
        "    X_Train_Norm = minmaxNorm(X_Train, _min, _max)\n",
        "    X_Valid_Norm = minmaxNorm(X_Valid, _min, _max)\n",
        "    all_distance = KNN_find_distance_each_data(X_Train_Norm, X_Valid_Norm)\n",
        "    all_sorted_target = KNN_find_sorted_target_each_data(Y_Train, all_distance)\n",
        "    K_list, error_list = KNNC_find_error_each_K(\n",
        "        Y_Valid, all_sorted_target, min_K, max_K)\n",
        "    best_K = KNN_find_best_K(K_list, error_list)\n",
        "    return best_K"
      ]
    },
    {
      "attachments": {},
      "cell_type": "markdown",
      "metadata": {},
      "source": [
        "### พยากรณ์\n"
      ]
    },
    {
      "attachments": {},
      "cell_type": "markdown",
      "metadata": {},
      "source": [
        "#### KNN-C\n"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 244,
      "metadata": {},
      "outputs": [],
      "source": [
        "def KNNC_predict(X_Train, Y_Train, X_Test, best_K):\n",
        "    _min = min4norm(X_Train)\n",
        "    _max = max4norm(X_Train)\n",
        "    X_Train_Norm = minmaxNorm(X_Train, _min, _max)\n",
        "    X_Test_Norm = minmaxNorm(X_Test, _min, _max)\n",
        "    all_distance = KNN_find_distance_each_data(X_Train, X_Test)\n",
        "    all_sorted_target = KNN_find_sorted_target_each_data(Y_Train, all_distance)\n",
        "    Yhat_Test = KNNC_find_class_each_data(all_sorted_target, best_K)\n",
        "    return Yhat_Test"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 245,
      "metadata": {
        "id": "G4_CGSTzCKaJ"
      },
      "outputs": [],
      "source": [
        "DataMatrix1 = df_train.values[:, :]\n",
        "DataMatrix2 = df_valid.values[:, :]\n",
        "DataMatrix3 = df_test.values\n",
        "\n",
        "D = 1\n",
        "X_train = DataMatrix1[:, D:]\n",
        "X_train = np.array(X_train, dtype=np.float32)\n",
        "\n",
        "X_valid = DataMatrix2[:, D:]\n",
        "X_valid = np.array(X_valid, dtype=np.float32)\n",
        "\n",
        "X_test = DataMatrix3[:, D:]\n",
        "X_test = np.array(X_test, dtype=np.float32)\n",
        "\n",
        "Y_train = DataMatrix1[:, :D]\n",
        "Y_valid = DataMatrix2[:, :D]\n",
        "Y_test = DataMatrix3[:, :D]"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 246,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "dmu7mQhnFLTc",
        "outputId": "11fe00d8-99c0-437a-cdc4-3365c8d14cd7"
      },
      "outputs": [
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "best_K คือ  39\n"
          ]
        },
        {
          "data": {
            "image/png": "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",
            "text/plain": [
              "<Figure size 640x480 with 1 Axes>"
            ]
          },
          "metadata": {},
          "output_type": "display_data"
        }
      ],
      "source": [
        "best_K = KNNC_fit(X_train, Y_train, X_valid, Y_valid, 1, 50)\n",
        "print('best_K คือ ', best_K)"
      ]
    },
    {
      "attachments": {},
      "cell_type": "markdown",
      "metadata": {
        "id": "xk06ovip1ixw"
      },
      "source": [
        "#### Make Prediction\n"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 247,
      "metadata": {},
      "outputs": [],
      "source": [
        "Yhat_Test = KNNC_predict(X_train, Y_train, X_test, best_K)"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 248,
      "metadata": {},
      "outputs": [],
      "source": [
        "error_Test = find_error_classification(Y_test, Yhat_Test)"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 249,
      "metadata": {},
      "outputs": [
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "มีค่า error เท่ากับ  71.42857142857143\n"
          ]
        }
      ],
      "source": [
        "print(\"มีค่า error เท่ากับ \", error_Test)"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 250,
      "metadata": {},
      "outputs": [
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "KNNC Accuracy: 0.2857142857142857\n"
          ]
        }
      ],
      "source": [
        "from sklearn.metrics import accuracy_score\n",
        "\n",
        "accuracy = accuracy_score(Y_test, Yhat_Test)\n",
        "\n",
        "print(f\"KNNC Accuracy: {accuracy}\")"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 251,
      "metadata": {},
      "outputs": [
        {
          "data": {
            "image/png": "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",
            "text/plain": [
              "<Figure size 640x480 with 2 Axes>"
            ]
          },
          "metadata": {},
          "output_type": "display_data"
        }
      ],
      "source": [
        "import matplotlib.pyplot as plt\n",
        "import seaborn as sns\n",
        "from sklearn.metrics import confusion_matrix\n",
        "\n",
        "\n",
        "# คำนวณ Confusion Matrix\n",
        "cm = confusion_matrix(Y_test, Yhat_Test)\n",
        "\n",
        "# แสดงผล Confusion Matrix ด้วย Heatmap\n",
        "plt.figure()\n",
        "sns.heatmap(cm, annot=True, cmap='Blues')\n",
        "plt.title('Confusion Matrix')\n",
        "plt.xlabel('Predicted Labels')\n",
        "plt.ylabel('True Labels')\n",
        "plt.show()"
      ]
    },
    {
      "attachments": {},
      "cell_type": "markdown",
      "metadata": {
        "id": "xJqnj1i7Q_ps"
      },
      "source": [
        "# Naive bayes\n"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 252,
      "metadata": {
        "id": "YHxec7RyQ_9g"
      },
      "outputs": [],
      "source": [
        "from sklearn.naive_bayes import GaussianNB\n",
        "from sklearn.metrics import accuracy_score"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 253,
      "metadata": {},
      "outputs": [],
      "source": [
        "X_train = df_train.iloc[:, 1:]\n",
        "Y_train = df_train.iloc[:, :1]\n",
        "X_valid = df_valid.iloc[:, 1:]\n",
        "Y_valid = df_valid.iloc[:, :1]\n",
        "X_test = df_test.iloc[:, 1:]\n",
        "Y_test = df_test.iloc[:, :1]"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 254,
      "metadata": {
        "id": "8ka8kni-RU-D"
      },
      "outputs": [],
      "source": [
        "# สร้างโมเดล Naive Bayes\n",
        "model = GaussianNB()\n",
        "\n",
        "# ฝึกโมเดลด้วยชุด Train\n",
        "model.fit(X_train,  Y_train.values.ravel())\n",
        "\n",
        "# ใช้โมเดลที่ฝึกแล้วทำนายคลาสของชุด test\n",
        "\n",
        "Y_pred = model.predict(X_test)"
      ]
    },
    {
      "attachments": {},
      "cell_type": "markdown",
      "metadata": {
        "id": "pglw-m6iVFrN"
      },
      "source": [
        "## Evaluating Model Naive bayes\n"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 255,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "pqfBiJ9KVF6Y",
        "outputId": "70d0bc2d-77dd-4ee5-b4f8-b8f422d83849"
      },
      "outputs": [
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "Naive Bayes Accuracy: 0.2857142857142857\n"
          ]
        }
      ],
      "source": [
        "# คำนวณค่าความแม่นยำของโมเดล\n",
        "accuracy = accuracy_score(Y_test, Y_pred)\n",
        "\n",
        "print(f\"Naive Bayes Accuracy: {accuracy}\")"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 256,
      "metadata": {},
      "outputs": [
        {
          "data": {
            "image/png": "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",
            "text/plain": [
              "<Figure size 640x480 with 2 Axes>"
            ]
          },
          "metadata": {},
          "output_type": "display_data"
        }
      ],
      "source": [
        "# คำนวณ Confusion Matrix\n",
        "cm = confusion_matrix(Y_test, Y_pred)\n",
        "\n",
        "# แสดงผล Confusion Matrix ด้วย Heatmap\n",
        "plt.figure()\n",
        "sns.heatmap(cm, annot=True, cmap='Blues')\n",
        "plt.title('Confusion Matrix')\n",
        "plt.xlabel('Predicted Labels')\n",
        "plt.ylabel('True Labels')\n",
        "plt.show()"
      ]
    },
    {
      "attachments": {},
      "cell_type": "markdown",
      "metadata": {
        "id": "k8Q8Q7AGTlD9"
      },
      "source": [
        "# Decision Tree\n"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 257,
      "metadata": {
        "id": "RquNdMN4ThkN"
      },
      "outputs": [],
      "source": [
        "from sklearn.tree import DecisionTreeClassifier\n",
        "\n",
        "\n",
        "# แยก features (X) และ target (Y) จาก df_train\n",
        "X_train = df_train.iloc[:, 1:]\n",
        "Y_train = df_train.iloc[:, :1]['pH']\n",
        "\n",
        "X_valid = df_valid.iloc[:, 1:]\n",
        "Y_valid = df_valid.iloc[:, :1]\n",
        "\n",
        "X_test = df_test.iloc[:, 1:]\n",
        "Y_test = df_test.iloc[:, :1]\n",
        "\n",
        "# สร้างโมเดล Decision Tree\n",
        "model = DecisionTreeClassifier()\n",
        "\n",
        "# ฝึกโมเดลด้วยชุด Train\n",
        "model.fit(X_train, Y_train)\n",
        "\n",
        "# ใช้โมเดลที่ฝึกแล้วทำนายคลาสของชุด Validation\n",
        "\n",
        "Y_pred = model.predict(X_test)"
      ]
    },
    {
      "attachments": {},
      "cell_type": "markdown",
      "metadata": {
        "id": "N_r7HsgZU8_9"
      },
      "source": [
        "# Evaluating Model Decision Tree\n"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 258,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "lgI2JCmKU-kW",
        "outputId": "c0a2a148-92fd-4a9a-9b54-dece5c66e0c8"
      },
      "outputs": [
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "Decision Tree Accuracy: 0.35714285714285715\n"
          ]
        }
      ],
      "source": [
        "# คำนวณค่าความแม่นยำของโมเดล\n",
        "accuracy = accuracy_score(Y_test, Y_pred)\n",
        "\n",
        "# แสดงผลลัพธ์\n",
        "print(f\"Decision Tree Accuracy: {accuracy}\")"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 259,
      "metadata": {},
      "outputs": [
        {
          "data": {
            "image/png": "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",
            "text/plain": [
              "<Figure size 640x480 with 2 Axes>"
            ]
          },
          "metadata": {},
          "output_type": "display_data"
        }
      ],
      "source": [
        "# คำนวณ Confusion Matrix\n",
        "cm = confusion_matrix(Y_test, Y_pred)\n",
        "\n",
        "# แสดงผล Confusion Matrix ด้วย Heatmap\n",
        "plt.figure()\n",
        "sns.heatmap(cm, annot=True, cmap='Blues')\n",
        "plt.title('Confusion Matrix')\n",
        "plt.xlabel('Predicted Labels')\n",
        "plt.ylabel('True Labels')\n",
        "plt.show()"
      ]
    }
  ],
  "metadata": {
    "accelerator": "TPU",
    "colab": {
      "provenance": []
    },
    "gpuClass": "standard",
    "kernelspec": {
      "display_name": "Python 3",
      "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": "3.11.4"
    }
  },
  "nbformat": 4,
  "nbformat_minor": 0
}