decisive-model-with-high-profit-potential-and-low-risk .ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Deskripsi Proyek\n",
    "Kamu bekerja di perusahaan tambang OilyGiant. Nah, tugasmu adalah mencari tempat terbaik untuk mengembangkan 200 titik sumur minyak baru.\n",
    "\n",
    "Untuk menyelesaikan tugas ini, kamu harus melakukan langkah-langkah berikut:\n",
    "Baca file dengan parameter yang dikumpulkan dari sumur-sumur minyak di wilayah terpilih: kualitas minyak dan volume cadangannya;\n",
    "\n",
    "Buat sebuah model untuk memprediksi volume cadangan di sumur-sumur baru;\n",
    "\n",
    "Pilih sumur minyak yang memiliki nilai estimasi tertinggi;\n",
    "\n",
    "Pilih wilayah dengan total keuntungan tertinggi dari sumur minyak yang dipilih.\n",
    "\n",
    "Kamu punya data sampel minyak dari tiga wilayah. Parameter setiap sumur minyak di wilayah tersebut sudah diketahui. Buat sebuah model yang akan membantumu memilih wilayah dengan margin keuntungan tertinggi. "
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Import Library"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 52,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Import Library yang dibutuhkan\n",
    "\n",
    "# Library untuk Data Processing\n",
    "import pandas as pd, numpy as np\n",
    "from sklearn.model_selection import train_test_split\n",
    "\n",
    "# Library untuk data Visualization\n",
    "import matplotlib.pyplot as plt\n",
    "import seaborn as sns\n",
    "\n",
    "# Library untuk Machine Learning\n",
    "from sklearn.linear_model import LinearRegression\n",
    "from sklearn.linear_model import Ridge\n",
    "\n",
    "# Library untuk metrics\n",
    "from sklearn.metrics import mean_squared_error"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Load dataset"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Memuat data\n",
    "data0 = pd.read_csv ('/datasets/geo_data_0.csv')\n",
    "data1 = pd.read_csv ('/datasets/geo_data_1.csv')\n",
    "data2 = pd.read_csv ('/datasets/geo_data_2.csv')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Geo data 0\n",
      "      id        f0        f1        f2     product\n",
      "0  txEyH  0.705745 -0.497823  1.221170  105.280062\n",
      "1  2acmU  1.334711 -0.340164  4.365080   73.037750\n",
      "2  409Wp  1.022732  0.151990  1.419926   85.265647\n",
      "3  iJLyR -0.032172  0.139033  2.978566  168.620776\n",
      "4  Xdl7t  1.988431  0.155413  4.751769  154.036647\n",
      "--------------------\n",
      "\n",
      "Geo data 1\n",
      "      id         f0         f1        f2     product\n",
      "0  kBEdx -15.001348  -8.276000 -0.005876    3.179103\n",
      "1  62mP7  14.272088  -3.475083  0.999183   26.953261\n",
      "2  vyE1P   6.263187  -5.948386  5.001160  134.766305\n",
      "3  KcrkZ -13.081196 -11.506057  4.999415  137.945408\n",
      "4  AHL4O  12.702195  -8.147433  5.004363  134.766305\n",
      "--------------------\n",
      "\n",
      "Geo data 2\n",
      "      id        f0        f1        f2     product\n",
      "0  fwXo0 -1.146987  0.963328 -0.828965   27.758673\n",
      "1  WJtFt  0.262778  0.269839 -2.530187   56.069697\n",
      "2  ovLUW  0.194587  0.289035 -5.586433   62.871910\n",
      "3  q6cA6  2.236060 -0.553760  0.930038  114.572842\n",
      "4  WPMUX -0.515993  1.716266  5.899011  149.600746\n",
      "--------------------\n",
      "\n"
     ]
    }
   ],
   "source": [
    "# Melakukan pengecekan isi data\n",
    "print('Geo data 0')\n",
    "print(data0.head(5))\n",
    "print('--------------------\\n')\n",
    "print('Geo data 1')\n",
    "print(data1.head(5))\n",
    "print('--------------------\\n')\n",
    "print('Geo data 2')\n",
    "print(data2.head(5))\n",
    "print('--------------------')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "<div class=\"alert-info\">\n",
    "    \n",
    "<b>Deskripsi data</b>\n",
    "\n",
    "id — ID unik sumur minyak\n",
    "\n",
    "f0, f1, f2 — tiga fitur titik wilayah (makna spesifiknya sebenarnya tidak penting, tetapi fitur itu sendiri sangat penting)\n",
    "\n",
    "product — volume cadangan di dalam sumur minyak (1 unit = 1.000 barel).\n",
    "\n",
    "</div>\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Data Quality Checking"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Geo data 0\n",
      "<class 'pandas.core.frame.DataFrame'>\n",
      "RangeIndex: 100000 entries, 0 to 99999\n",
      "Data columns (total 5 columns):\n",
      " #   Column   Non-Null Count   Dtype  \n",
      "---  ------   --------------   -----  \n",
      " 0   id       100000 non-null  object \n",
      " 1   f0       100000 non-null  float64\n",
      " 2   f1       100000 non-null  float64\n",
      " 3   f2       100000 non-null  float64\n",
      " 4   product  100000 non-null  float64\n",
      "dtypes: float64(4), object(1)\n",
      "memory usage: 3.8+ MB\n",
      "------------------------------------\n",
      "\n",
      "Geo data 1\n",
      "<class 'pandas.core.frame.DataFrame'>\n",
      "RangeIndex: 100000 entries, 0 to 99999\n",
      "Data columns (total 5 columns):\n",
      " #   Column   Non-Null Count   Dtype  \n",
      "---  ------   --------------   -----  \n",
      " 0   id       100000 non-null  object \n",
      " 1   f0       100000 non-null  float64\n",
      " 2   f1       100000 non-null  float64\n",
      " 3   f2       100000 non-null  float64\n",
      " 4   product  100000 non-null  float64\n",
      "dtypes: float64(4), object(1)\n",
      "memory usage: 3.8+ MB\n",
      "------------------------------------\n",
      "\n",
      "Geo data 2\n",
      "<class 'pandas.core.frame.DataFrame'>\n",
      "RangeIndex: 100000 entries, 0 to 99999\n",
      "Data columns (total 5 columns):\n",
      " #   Column   Non-Null Count   Dtype  \n",
      "---  ------   --------------   -----  \n",
      " 0   id       100000 non-null  object \n",
      " 1   f0       100000 non-null  float64\n",
      " 2   f1       100000 non-null  float64\n",
      " 3   f2       100000 non-null  float64\n",
      " 4   product  100000 non-null  float64\n",
      "dtypes: float64(4), object(1)\n",
      "memory usage: 3.8+ MB\n",
      "------------------------------------\n"
     ]
    }
   ],
   "source": [
    "# Melakukan pengecekan informasi data\n",
    "print('Geo data 0')\n",
    "data0.info()\n",
    "print('------------------------------------\\n')\n",
    "print('Geo data 1')\n",
    "data1.info()\n",
    "print('------------------------------------\\n')\n",
    "print('Geo data 2')\n",
    "data2.info()\n",
    "print('------------------------------------')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "<div class=\"alert-info\">\n",
    "<b>Note</b>\n",
    "\n",
    "Tidak terdapat data yang hilang pada semua data geo.\n",
    "    \n",
    "    \n",
    "\n",
    "</div>"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Geo data 0 \n",
      " 0\n",
      "------------------------------------\n",
      "\n",
      "Geo data 1 \n",
      " 0\n",
      "------------------------------------\n",
      "\n",
      "Geo data 2 \n",
      " 0\n",
      "------------------------------------\n"
     ]
    }
   ],
   "source": [
    "# Cek duplikasi data\n",
    "print('Geo data 0 \\n', data0.duplicated().sum())\n",
    "print('------------------------------------\\n')\n",
    "print('Geo data 1 \\n', data1.duplicated().sum())\n",
    "print('------------------------------------\\n')\n",
    "print('Geo data 2 \\n', data2.duplicated().sum())\n",
    "print('------------------------------------')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "<div class=\"alert-info\">\n",
    "<b>Note</b>\n",
    "\n",
    "Tidak terdapat data duplikat pada semua data geo.\n",
    "\n",
    "    \n",
    "    \n",
    "\n",
    "</div>"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Analisis data eksploratif (EDA) "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 432x288 with 2 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "# Melakukan pengecekan Korelasi antara fiture dan target.\n",
    "\n",
    "corr0 = data0.corr()\n",
    "plt.title('Correlation di Geo data 0')\n",
    "sns.heatmap(corr0, annot=True)\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "<div class=\"alert-info\">\n",
    "<b>Note</b>\n",
    "\n",
    "- F0 dan F1, Memiliki korelasi yang lemah terhadap produk / Target\n",
    "- F2, Memiliki Korelasi sedang terhadap produk / Target\n",
    "    \n",
    "    \n",
    "\n",
    "</div>"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 432x288 with 2 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "# Melakukan pengecekan Korelasi antara fiture dan target.\n",
    "\n",
    "corr1 = data1.corr()\n",
    "plt.title('Correlation di Geo data 1')\n",
    "sns.heatmap(corr1, annot=True)\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "<div class=\"alert-info\">\n",
    "<b>Note</b>\n",
    "\n",
    "- F0 dan F1, Memiliki korelasi yang lemah terhadap produk / Target\n",
    "- F2, Memiliki Korelasi yang sangat kuat terhadap produk / Target\n",
    "    \n",
    "    \n",
    "\n",
    "</div>"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 432x288 with 2 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "# Melakukan pengecekan Korelasi antara fiture dan target.\n",
    "\n",
    "corr2 = data2.corr()\n",
    "plt.title('Correlation di Geo data 2')\n",
    "sns.heatmap(corr2, annot=True)\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "<div class=\"alert-info\">\n",
    "<b>Note</b>\n",
    "\n",
    "- F0 dan F1, Memiliki korelasi yang lemah terhadap produk / Target\n",
    "- F2, Memiliki Korelasi sedang terhadap produk / Target\n",
    "    \n",
    "    \n",
    "\n",
    "</div>"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Model Creation"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 32,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Membuat function Mean Absolute Percentage Error (MAPE)\n",
    "def mape(y_actual, y_predict):\n",
    "    metric = pd.Series(abs(y_actual - y_predict) / y_actual).replace(np.inf, 0).mean() * 100\n",
    "    return metric"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 106,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Membuat function model, hasil RSME dan MAPE\n",
    "def create_model(dataframe, model, data_ke):\n",
    "    # Pisahkan X dan y\n",
    "    feature = dataframe[['f0', 'f1', 'f2']]\n",
    "    target = dataframe['product']\n",
    "\n",
    "    # Train Test Split\n",
    "    lr = LinearRegression()\n",
    "    X_train, X_test, y_train, y_test = train_test_split(feature, target, test_size=0.25, random_state=10)\n",
    "\n",
    "\n",
    "    # Prediksi Data\n",
    "    model.fit(X_train, y_train)\n",
    "    y_pred_train = model.predict(X_train)\n",
    "    y_pred_test  = model.predict(X_test)\n",
    "\n",
    "    # Evaluasi Metrics dari Model\n",
    "    rmse_train = np.sqrt(mean_squared_error(y_train, y_pred_train))\n",
    "    rmse_test = np.sqrt(mean_squared_error(y_test, y_pred_test))\n",
    "\n",
    "    # Print RMSE\n",
    "    print('Geo Data ke', data_ke, '----------------------')\n",
    "    print('RMSE Train   :', round(rmse_train, 2))\n",
    "    print('RMSE Test    :', round(rmse_test, 2))\n",
    "    # Print MAPE\n",
    "    print('MAPE Train   :', round(mape(y_train, y_pred_train), 2), '%')\n",
    "    print('MAPE Test    :', round(mape(y_test, y_pred_test), 2), '%')\n",
    "\n",
    "    # Rata2 Volume Minyak yang terprediksi\n",
    "    mean_volume = y_pred_test.mean()\n",
    "    print('Volume Minyak:', round(mean_volume, 2))\n",
    "\n",
    "    result = pd.DataFrame({'actual': y_test, 'predict': y_pred_test})\n",
    "\n",
    "    return result"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### LinearRegression\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 63,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Membuat Model Linear Regression\n",
    "lr = LinearRegression()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 107,
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Geo Data ke 0 ----------------------\n",
      "RMSE Train   : 37.64\n",
      "RMSE Test    : 37.85\n",
      "MAPE Train   : 147.33 %\n",
      "MAPE Test    : 245.61 %\n",
      "Volume Minyak: 92.41\n",
      "Geo Data ke 1 ----------------------\n",
      "RMSE Train   : 0.89\n",
      "RMSE Test    : 0.89\n",
      "MAPE Train   : 2.86 %\n",
      "MAPE Test    : 2.82 %\n",
      "Volume Minyak: 68.83\n",
      "Geo Data ke 2 ----------------------\n",
      "RMSE Train   : 40.06\n",
      "RMSE Test    : 40.05\n",
      "MAPE Train   : 200.68 %\n",
      "MAPE Test    : 237.09 %\n",
      "Volume Minyak: 94.96\n"
     ]
    }
   ],
   "source": [
    "# Predict data ke 0\n",
    "result0 = create_model(dataframe=data0, model=lr, data_ke=0)\n",
    "\n",
    "# Predict data ke 1\n",
    "result1 = create_model(dataframe=data1, model=lr, data_ke=1)\n",
    "\n",
    "# Predict data ke 2\n",
    "result2 = create_model(dataframe=data2, model=lr, data_ke=2)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "<div class=\"alert-info\">\n",
    "<b>Insight</b>\n",
    "\n",
    "- rata-rata Volume minyak terbanyak terdapat pada Geo Data 2, Namun memiliki RSME yang paling besar.\n",
    "- RMSE yang paling kecil terdapat pada Geo data 1, namun memiliki rata-rata volume minyak terkecil.\n",
    "    \n",
    "    \n",
    "\n",
    "</div>"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Ridge"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 60,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Membuat Model Ridge\n",
    "ridge = Ridge()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 61,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Data ke 0 ----------------------\n",
      "RMSE Train   : 37.64\n",
      "RMSE Test    : 37.85\n",
      "MAPE Train   : 147.33 %\n",
      "MAPE Test    : 245.61 %\n",
      "Volume Minyak: 92.41\n",
      "Data ke 1 ----------------------\n",
      "RMSE Train   : 0.89\n",
      "RMSE Test    : 0.89\n",
      "MAPE Train   : 2.86 %\n",
      "MAPE Test    : 2.82 %\n",
      "Volume Minyak: 68.83\n",
      "Data ke 2 ----------------------\n",
      "RMSE Train   : 40.06\n",
      "RMSE Test    : 40.05\n",
      "MAPE Train   : 200.68 %\n",
      "MAPE Test    : 237.09 %\n",
      "Volume Minyak: 94.96\n"
     ]
    }
   ],
   "source": [
    "# Predict data ke 0\n",
    "result0_ridge = create_model(dataframe=data0, model=ridge, data_ke=0)\n",
    "\n",
    "# Predict data ke 1\n",
    "result1_ridge = create_model(dataframe=data1, model=ridge, data_ke=1)\n",
    "\n",
    "# Predict data ke 2\n",
    "result2_ridge = create_model(dataframe=data2, model=ridge, data_ke=2)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "<div class=\"alert-info\">\n",
    "<b>Insight</b>\n",
    "\n",
    "Hasilnya sama persis dengan model Linier Regression    \n",
    "- rata-rata Volume minyak terbanyak terdapat pada Geo Data 2, Namun memiliki RSME yang paling besar.\n",
    "- RMSE yang paling kecil terdapat pada Geo data 1, namun memiliki rata-rata volume minyak terkecil.\n",
    "    \n",
    "    \n",
    "\n",
    "</div>"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Model Valuation"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 65,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Benchmark\n",
    "anggaran    = 100000000\n",
    "pendapatan  = 4500\n",
    "sumur       = 200"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Perhitungan Keuntungan / Laba"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 67,
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Minimum Pendapatan per Sumur       : $ 500000.0\n",
      "Minimum Volume Minyak yg dihasilkan: 111.11  ribu barrels\n"
     ]
    }
   ],
   "source": [
    "cost_sumur  = anggaran / sumur\n",
    "min_barrel  = cost_sumur / pendapatan\n",
    "print('Minimum Pendapatan per Sumur       : $', round(cost_sumur, 2))\n",
    "print('Minimum Volume Minyak yg dihasilkan:', round(min_barrel, 2), ' ribu barrels')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 116,
   "metadata": {},
   "outputs": [],
   "source": [
    "def revenue(dataframe):\n",
    "    pred_sorted = dataframe.sort_values(by='predict', ascending=False).head(200)\n",
    "    selected = pred_sorted['actual'].sum()\n",
    "    revenue = selected * 4500 - 100000000\n",
    "    return revenue"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 117,
   "metadata": {},
   "outputs": [],
   "source": [
    "def bootstrap_revenue(dataframe, bootstrap=1000):\n",
    "    state = np.random.RandomState(12345)\n",
    "    values = []\n",
    "    \n",
    "    for i in range(bootstrap):\n",
    "        target_sample = dataframe.sample(n=500, replace=True, random_state=state)\n",
    "        target_sample_revenue = revenue(target_sample)\n",
    "        values.append(target_sample_revenue)\n",
    "    \n",
    "    return pd.Series(values)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 124,
   "metadata": {},
   "outputs": [],
   "source": [
    "def profit(data, percobaan=1000):\n",
    "    boot = bootstrap_revenue(dataframe=data, bootstrap=percobaan)\n",
    "    confidence_inteval = 0.95\n",
    "    alpha = 1 - confidence_inteval\n",
    "    min = int(boot.quantile(alpha/2))\n",
    "    max = int(boot.quantile(1-alpha/2))\n",
    "    avg = int(boot.mean())\n",
    "    risk = round((boot < 0).mean() * 100, 2)\n",
    "    \n",
    "    print(f'Range Confidence : $({min}) - $({max})')\n",
    "    print(f'Rata-rata Profit : ${avg}')\n",
    "    print(f'Risiko Kerugian  : {risk}%')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 125,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Result Geo data 0\n",
      "Range Confidence : $(-1302336) - $(8935740)\n",
      "Rata-rata Profit : $3773683\n",
      "Risiko Kerugian  : 8.1%\n",
      "------------------- \n",
      "\n",
      "Result Geo data 1\n",
      "Range Confidence : $(907582) - $(8829726)\n",
      "Rata-rata Profit : $4732811\n",
      "Risiko Kerugian  : 1.5%\n",
      "------------------- \n",
      "\n",
      "Result Geo data 2\n",
      "Range Confidence : $(-1734909) - $(9166356)\n",
      "Rata-rata Profit : $3603957\n",
      "Risiko Kerugian  : 9.0%\n",
      "-------------------\n"
     ]
    }
   ],
   "source": [
    "print('Result Geo data 0')\n",
    "profit(data=result0_lr)\n",
    "print('------------------- \\n')\n",
    "\n",
    "print('Result Geo data 1')\n",
    "profit(data=result1_lr)\n",
    "print('------------------- \\n')\n",
    "\n",
    "\n",
    "print('Result Geo data 2')\n",
    "profit(data=result2_lr)\n",
    "print('-------------------')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "<div class=\"alert-info\">\n",
    "<b>Insight</b>\n",
    "\n",
    "Hasilnya sama persis dengan model Linier Regression    \n",
    "- rata-rata Volume minyak terbanyak terdapat pada Geo Data 2, Namun memiliki RSME yang paling besar.\n",
    "- RMSE yang paling kecil terdapat pada Geo data 1, namun memiliki rata-rata volume minyak terkecil.\n",
    "    \n",
    "    \n",
    "\n",
    "</div>"
   ]
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3 (ipykernel)",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.9.19"
  },
  "toc": {
   "base_numbering": 1,
   "nav_menu": {},
   "number_sections": true,
   "sideBar": true,
   "skip_h1_title": true,
   "title_cell": "Table of Contents",
   "title_sidebar": "Contents",
   "toc_cell": false,
   "toc_position": {},
   "toc_section_display": true,
   "toc_window_display": false
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}