29 - Random Forest for Regression - Prediction of Continuous Well Logs-Copy1.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "id": "e3dcafed",
   "metadata": {},
   "source": [
    "# Random Forest for Predicting Continuous Well Measurements"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "33025840",
   "metadata": {},
   "source": [
    "## Importing Libraries and Loading Data"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "id": "731ac85c",
   "metadata": {},
   "outputs": [],
   "source": [
    "import pandas as pd\n",
    "import matplotlib.pyplot as plt"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "id": "c2fb85af",
   "metadata": {},
   "outputs": [],
   "source": [
    "df = pd.read_csv('Data/Volve/volve_wells.csv', usecols=['WELL', 'DEPTH', 'RHOB', 'GR', 'NPHI', 'PEF', 'DT'])"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "1c86690f",
   "metadata": {},
   "source": [
    "## Create Training, Testing and Validation Datasets"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "01aade55",
   "metadata": {},
   "source": [
    "Our dataset should have 4 wells within it. We can confirm this by calling upon the `unique()` function"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "id": "7c6c58be",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array(['15/9-F-11 B', '15/9-F-11 A', '15/9-F-1 B', '15/9-F-1 A'],\n",
       "      dtype=object)"
      ]
     },
     "execution_count": 3,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df['WELL'].unique()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "0607c9d1",
   "metadata": {},
   "source": [
    "As we are using measurements taken from multiple wells, one way to split our data into training and testing is to set aside a single well (blind test well) which will be used to see how our model performs on unseen data."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "id": "c7114040",
   "metadata": {},
   "outputs": [],
   "source": [
    "# Training Wells\n",
    "training_wells = ['15/9-F-11 B', '15/9-F-11 A', '15/9-F-1 A']\n",
    "\n",
    "# Test Well\n",
    "test_well = ['15/9-F-1 B']"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "dfd658cc",
   "metadata": {},
   "source": [
    "\"Extract\" the data from the main dataframe using the well lists above"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "id": "6320823a",
   "metadata": {},
   "outputs": [],
   "source": [
    "train_val_df = df[df['WELL'].isin(training_wells)].copy()\n",
    "test_df = df[df['WELL'].isin(test_well)].copy()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "id": "37d705af",
   "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>DEPTH</th>\n",
       "      <th>DT</th>\n",
       "      <th>GR</th>\n",
       "      <th>NPHI</th>\n",
       "      <th>PEF</th>\n",
       "      <th>RHOB</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>count</th>\n",
       "      <td>116914.000000</td>\n",
       "      <td>21699.000000</td>\n",
       "      <td>115933.000000</td>\n",
       "      <td>37587.000000</td>\n",
       "      <td>37668.000000</td>\n",
       "      <td>37668.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>mean</th>\n",
       "      <td>2154.233438</td>\n",
       "      <td>77.252247</td>\n",
       "      <td>51.823119</td>\n",
       "      <td>0.174302</td>\n",
       "      <td>6.450603</td>\n",
       "      <td>2.443072</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>std</th>\n",
       "      <td>1180.976133</td>\n",
       "      <td>14.350893</td>\n",
       "      <td>37.606884</td>\n",
       "      <td>0.085660</td>\n",
       "      <td>1.478121</td>\n",
       "      <td>0.166466</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>min</th>\n",
       "      <td>145.900000</td>\n",
       "      <td>53.165000</td>\n",
       "      <td>0.149100</td>\n",
       "      <td>0.010000</td>\n",
       "      <td>3.647000</td>\n",
       "      <td>1.627000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>25%</th>\n",
       "      <td>1148.525000</td>\n",
       "      <td>66.854450</td>\n",
       "      <td>22.126100</td>\n",
       "      <td>0.115000</td>\n",
       "      <td>5.078850</td>\n",
       "      <td>2.276000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>50%</th>\n",
       "      <td>2122.800000</td>\n",
       "      <td>72.724000</td>\n",
       "      <td>52.217000</td>\n",
       "      <td>0.163000</td>\n",
       "      <td>6.548700</td>\n",
       "      <td>2.501000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>75%</th>\n",
       "      <td>3097.100000</td>\n",
       "      <td>86.132300</td>\n",
       "      <td>74.201000</td>\n",
       "      <td>0.212100</td>\n",
       "      <td>7.728625</td>\n",
       "      <td>2.577000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>max</th>\n",
       "      <td>4770.200000</td>\n",
       "      <td>126.827000</td>\n",
       "      <td>1124.403000</td>\n",
       "      <td>0.593200</td>\n",
       "      <td>13.841000</td>\n",
       "      <td>3.090000</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "               DEPTH            DT             GR          NPHI           PEF  \\\n",
       "count  116914.000000  21699.000000  115933.000000  37587.000000  37668.000000   \n",
       "mean     2154.233438     77.252247      51.823119      0.174302      6.450603   \n",
       "std      1180.976133     14.350893      37.606884      0.085660      1.478121   \n",
       "min       145.900000     53.165000       0.149100      0.010000      3.647000   \n",
       "25%      1148.525000     66.854450      22.126100      0.115000      5.078850   \n",
       "50%      2122.800000     72.724000      52.217000      0.163000      6.548700   \n",
       "75%      3097.100000     86.132300      74.201000      0.212100      7.728625   \n",
       "max      4770.200000    126.827000    1124.403000      0.593200     13.841000   \n",
       "\n",
       "               RHOB  \n",
       "count  37668.000000  \n",
       "mean       2.443072  \n",
       "std        0.166466  \n",
       "min        1.627000  \n",
       "25%        2.276000  \n",
       "50%        2.501000  \n",
       "75%        2.577000  \n",
       "max        3.090000  "
      ]
     },
     "execution_count": 6,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "train_val_df.describe()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "id": "fc094c52",
   "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>DEPTH</th>\n",
       "      <th>DT</th>\n",
       "      <th>GR</th>\n",
       "      <th>NPHI</th>\n",
       "      <th>PEF</th>\n",
       "      <th>RHOB</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>count</th>\n",
       "      <td>33191.000000</td>\n",
       "      <td>4262.000000</td>\n",
       "      <td>32498.000000</td>\n",
       "      <td>3413.000000</td>\n",
       "      <td>3441.000000</td>\n",
       "      <td>3441.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>mean</th>\n",
       "      <td>1805.400000</td>\n",
       "      <td>80.380006</td>\n",
       "      <td>56.864115</td>\n",
       "      <td>0.203836</td>\n",
       "      <td>6.445014</td>\n",
       "      <td>2.453695</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>std</th>\n",
       "      <td>958.156073</td>\n",
       "      <td>14.698333</td>\n",
       "      <td>35.935409</td>\n",
       "      <td>0.095991</td>\n",
       "      <td>0.811407</td>\n",
       "      <td>0.129990</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>min</th>\n",
       "      <td>145.900000</td>\n",
       "      <td>54.928300</td>\n",
       "      <td>0.149300</td>\n",
       "      <td>0.059500</td>\n",
       "      <td>4.729900</td>\n",
       "      <td>2.111800</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>25%</th>\n",
       "      <td>975.650000</td>\n",
       "      <td>71.513225</td>\n",
       "      <td>29.481175</td>\n",
       "      <td>0.146900</td>\n",
       "      <td>5.938300</td>\n",
       "      <td>2.379200</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>50%</th>\n",
       "      <td>1805.400000</td>\n",
       "      <td>76.295350</td>\n",
       "      <td>58.005700</td>\n",
       "      <td>0.172700</td>\n",
       "      <td>6.352900</td>\n",
       "      <td>2.482100</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>75%</th>\n",
       "      <td>2635.150000</td>\n",
       "      <td>85.365475</td>\n",
       "      <td>79.247200</td>\n",
       "      <td>0.222300</td>\n",
       "      <td>6.872600</td>\n",
       "      <td>2.536700</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>max</th>\n",
       "      <td>3464.900000</td>\n",
       "      <td>125.982700</td>\n",
       "      <td>297.767300</td>\n",
       "      <td>0.557600</td>\n",
       "      <td>10.987600</td>\n",
       "      <td>3.051700</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "              DEPTH           DT            GR         NPHI          PEF  \\\n",
       "count  33191.000000  4262.000000  32498.000000  3413.000000  3441.000000   \n",
       "mean    1805.400000    80.380006     56.864115     0.203836     6.445014   \n",
       "std      958.156073    14.698333     35.935409     0.095991     0.811407   \n",
       "min      145.900000    54.928300      0.149300     0.059500     4.729900   \n",
       "25%      975.650000    71.513225     29.481175     0.146900     5.938300   \n",
       "50%     1805.400000    76.295350     58.005700     0.172700     6.352900   \n",
       "75%     2635.150000    85.365475     79.247200     0.222300     6.872600   \n",
       "max     3464.900000   125.982700    297.767300     0.557600    10.987600   \n",
       "\n",
       "              RHOB  \n",
       "count  3441.000000  \n",
       "mean      2.453695  \n",
       "std       0.129990  \n",
       "min       2.111800  \n",
       "25%       2.379200  \n",
       "50%       2.482100  \n",
       "75%       2.536700  \n",
       "max       3.051700  "
      ]
     },
     "execution_count": 7,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "test_df.describe()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "df9ab50f",
   "metadata": {},
   "source": [
    "### Remove NaN Values From Dataframe\n",
    "Removing missing values from the dataframe is one way to deal with them, however, doing so reduces the amount of training data you have available. Other methods can be used to infill the NaNs with sensible values."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "id": "d3a9d82c",
   "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>DEPTH</th>\n",
       "      <th>DT</th>\n",
       "      <th>GR</th>\n",
       "      <th>NPHI</th>\n",
       "      <th>PEF</th>\n",
       "      <th>RHOB</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>count</th>\n",
       "      <td>21688.000000</td>\n",
       "      <td>21688.000000</td>\n",
       "      <td>21688.000000</td>\n",
       "      <td>21688.000000</td>\n",
       "      <td>21688.000000</td>\n",
       "      <td>21688.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>mean</th>\n",
       "      <td>3141.098875</td>\n",
       "      <td>77.235857</td>\n",
       "      <td>39.803246</td>\n",
       "      <td>0.166648</td>\n",
       "      <td>7.093603</td>\n",
       "      <td>2.475232</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>std</th>\n",
       "      <td>314.723749</td>\n",
       "      <td>14.336048</td>\n",
       "      <td>57.907158</td>\n",
       "      <td>0.099200</td>\n",
       "      <td>1.188313</td>\n",
       "      <td>0.147635</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>min</th>\n",
       "      <td>2577.000000</td>\n",
       "      <td>53.165000</td>\n",
       "      <td>0.852000</td>\n",
       "      <td>0.010000</td>\n",
       "      <td>4.297800</td>\n",
       "      <td>1.980600</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>25%</th>\n",
       "      <td>2869.475000</td>\n",
       "      <td>66.849300</td>\n",
       "      <td>9.416350</td>\n",
       "      <td>0.096000</td>\n",
       "      <td>6.218475</td>\n",
       "      <td>2.379000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>50%</th>\n",
       "      <td>3140.550000</td>\n",
       "      <td>72.720750</td>\n",
       "      <td>27.552000</td>\n",
       "      <td>0.136000</td>\n",
       "      <td>7.487700</td>\n",
       "      <td>2.533000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>75%</th>\n",
       "      <td>3411.625000</td>\n",
       "      <td>86.093800</td>\n",
       "      <td>44.877425</td>\n",
       "      <td>0.217200</td>\n",
       "      <td>8.001000</td>\n",
       "      <td>2.581400</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>max</th>\n",
       "      <td>3723.300000</td>\n",
       "      <td>126.827000</td>\n",
       "      <td>1124.403000</td>\n",
       "      <td>0.593200</td>\n",
       "      <td>13.841000</td>\n",
       "      <td>3.025000</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "              DEPTH            DT            GR          NPHI           PEF  \\\n",
       "count  21688.000000  21688.000000  21688.000000  21688.000000  21688.000000   \n",
       "mean    3141.098875     77.235857     39.803246      0.166648      7.093603   \n",
       "std      314.723749     14.336048     57.907158      0.099200      1.188313   \n",
       "min     2577.000000     53.165000      0.852000      0.010000      4.297800   \n",
       "25%     2869.475000     66.849300      9.416350      0.096000      6.218475   \n",
       "50%     3140.550000     72.720750     27.552000      0.136000      7.487700   \n",
       "75%     3411.625000     86.093800     44.877425      0.217200      8.001000   \n",
       "max     3723.300000    126.827000   1124.403000      0.593200     13.841000   \n",
       "\n",
       "               RHOB  \n",
       "count  21688.000000  \n",
       "mean       2.475232  \n",
       "std        0.147635  \n",
       "min        1.980600  \n",
       "25%        2.379000  \n",
       "50%        2.533000  \n",
       "75%        2.581400  \n",
       "max        3.025000  "
      ]
     },
     "execution_count": 8,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "train_val_df.dropna(inplace=True)\n",
    "test_df.dropna(inplace=True)\n",
    "train_val_df.describe()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "4e55a73d",
   "metadata": {},
   "source": [
    "## Implementing the Random Forest Model"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "id": "fa08e1cc",
   "metadata": {},
   "outputs": [],
   "source": [
    "from sklearn.model_selection import train_test_split\n",
    "from sklearn import metrics\n",
    "from sklearn.ensemble import RandomForestRegressor"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "b18c90d6",
   "metadata": {},
   "source": [
    "### Selecting Training and Target Features"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "id": "fd8dc2b5",
   "metadata": {},
   "outputs": [],
   "source": [
    "X = train_val_df[['RHOB', 'GR', 'NPHI', 'PEF']]\n",
    "y = train_val_df['DT']"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "51ec1cf5",
   "metadata": {},
   "source": [
    "Note that the name test used here is commonly used within machine learning. In this case the variables X_test and y_test are our validation data. In other words it is used to help tune our model. "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "id": "fb9b92c0",
   "metadata": {},
   "outputs": [],
   "source": [
    "X_train, X_val, y_train, y_val = train_test_split(X, y, test_size=0.2)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "76c0fe11",
   "metadata": {},
   "source": [
    "Checking the shapes of X_train and X_test to make sure they have been split correctly."
   ]
  },
  {
   "cell_type": "markdown",
   "id": "6508af88",
   "metadata": {},
   "source": [
    "### Building the Model"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "id": "4d6904f6",
   "metadata": {},
   "outputs": [],
   "source": [
    "regr = RandomForestRegressor()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "id": "4a37dd73",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "RandomForestRegressor()"
      ]
     },
     "execution_count": 15,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "regr.fit(X_train, y_train)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "id": "8736e39b",
   "metadata": {},
   "outputs": [],
   "source": [
    "y_pred = regr.predict(X_val)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "f4443b15",
   "metadata": {},
   "source": [
    "### Check the Prediction Results"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "id": "f11706f0",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "1.684142094744122"
      ]
     },
     "execution_count": 17,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "metrics.mean_absolute_error(y_val, y_pred)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "id": "b225f0c1",
   "metadata": {},
   "outputs": [],
   "source": [
    "mse = metrics.mean_squared_error(y_val, y_pred)\n",
    "rmse = mse**0.5 "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "id": "7387a999",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "3.066405718913411"
      ]
     },
     "execution_count": 19,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "rmse"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "328fca04",
   "metadata": {},
   "source": [
    "Simple metrics like above are a nice way to see how a model has performed, but you should always check the actual data. \n",
    "\n",
    "In the plot below, we are comparing the real data against the predicted data."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "id": "6c5639e7",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "[<matplotlib.lines.Line2D at 0x12fa88520>]"
      ]
     },
     "execution_count": 21,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.scatter(y_val, y_pred)\n",
    "plt.xlim(40, 140)\n",
    "plt.ylim(40, 140)\n",
    "plt.ylabel('Predicted DT')\n",
    "plt.xlabel('Actual DT')\n",
    "plt.plot([40,140], [40,140], 'black') #1 to 1 line"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "742e3edd",
   "metadata": {},
   "source": [
    "## Test Well Prediction\n",
    "Once the model has been fine tuned, we can apply it to our blind test well and see how it performs."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "id": "9ffe8a16",
   "metadata": {},
   "outputs": [],
   "source": [
    "test_well_x = test_df[['RHOB', 'GR', 'NPHI', 'PEF']]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "id": "af0d83da",
   "metadata": {},
   "outputs": [],
   "source": [
    "test_df['TEST_DT'] = regr.predict(test_well_x)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 25,
   "id": "71e6109f",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "[<matplotlib.lines.Line2D at 0x12fb4cf70>]"
      ]
     },
     "execution_count": 25,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.scatter(test_df['DT'], test_df['TEST_DT'])\n",
    "plt.xlim(40, 140)\n",
    "plt.ylim(40, 140)\n",
    "plt.ylabel('Predicted DT')\n",
    "plt.xlabel('Actual DT')\n",
    "plt.plot([40,140], [40,140], 'black') #1 to 1 line"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 27,
   "id": "ccb99db5",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1080x360 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.figure(figsize=(15, 5))\n",
    "plt.plot(test_df['DEPTH'], test_df['DT'], label='Actual DT')\n",
    "plt.plot(test_df['DEPTH'], test_df['TEST_DT'], label='Predicted DT')\n",
    "plt.xlabel('Depth (m)')\n",
    "plt.ylabel('DT')\n",
    "plt.ylim(40, 140)\n",
    "plt.legend()\n",
    "plt.grid()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "1859d8ec",
   "metadata": {},
   "outputs": [],
   "source": []
  }
 ],
 "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": "3.10.3"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 5
}