{
 "cells": [
  {
   "cell_type": "markdown",
   "id": "20363879-23d9-4ef7-a6ef-c46858651ef3",
   "metadata": {},
   "source": [
    "# 🌊 Predicting Significant Wave Heights for Ocean Exploration (70/100)\n",
    "\n",
    "\n",
    "Welcome to MSC-AI Consulting! You are leading a consulting team in an important project with a top client, the Global Ocean Exploration Initiative (GOEI). This client focuses on understanding the dynamic nature of oceanic waves and their potential impact on marine operations.\n",
    "\n",
    "\n",
    "## 📄 Client’s Brief\n",
    "GOEI has provided you with a dataset that records oceanic and environmental conditions, and they would like you to predict significant wave heights (`Hsig`) based on these conditions. Accurate wave height predictions are crucial for planning maritime operations and ensuring safety.\n",
    "\n",
    "\n",
    "## Understanding the Data\n",
    "\n",
    "The data provided in `Hs.csv` includes several features that capture environmental and oceanic conditions related to wave height. Here are the key columns in the dataset:\n",
    "\n",
    "- `Hsig`: Significant wave height, measured in meters, which serves as the target variable for your regression model. Predicting this feature accurately is essential for understanding ocean conditions.\n",
    "- `Temperature`: Represents the water temperature, which can be analyzed in its continuous form or transformed into categories (e.g., Low, Moderate, High) to capture potential temperature bands.\n",
    "- `Wind Speed`: The speed of the wind measured in meters per second (m/s). Wind speed is a critical factor that influences wave heights and ocean conditions.\n",
    "- `Wave Direction (Dir)`: The direction from which the waves are approaching, measured in degrees (0° - 360°). This feature can help reveal patterns associated with wave height depending on the wave’s origin.\n",
    "- `Depth`: The depth of the water at the measurement location, measured in meters. Depth provides context for interpreting oceanic conditions at different levels below the surface.\n",
    "- `X-Windv` and `Y-Windv`: Components of wind velocity along the x-axis and y-axis, which capture directional wind forces.\n",
    "- `Season`: Represents the seasonal classification (e.g., Winter, Spring, Summer, Fall) for each measurement, allowing you to capture potential seasonal effects on wave heights.\n",
    "- `Wind_Dir_Category`: A categorical version of Wave Direction, grouped into broader directional categories (e.g., North, East, South, West) for simplified analysis of directional effects on wave height.\n",
    "\n",
    "# Part 1: Data Preprocessing\n",
    "\n",
    "## Objective\n",
    "Your primary goal is to predict Hsig (significant wave height).\n",
    "\n",
    "## Train-Test Split\n",
    "Begin by splitting the dataset into 70% for training and 30% for testing. Given the structure of this project, there’s no need for a separate validation set, as GOEI wants you to maximize training data usage through cross-validation.\n",
    "\n",
    "## Data Preprocessing\n",
    "- Use insights from your EDA to inform preprocessing steps. While specific steps are up to you, standard preprocessing tasks should be implemented.\n",
    "- The client has confirmed that no outliers are present in the dataset, so outlier handling is unnecessary.\n",
    "\n",
    "## Feature Engineering and Selection\n",
    "In addition to all standard steps (e.g., addressing collinearity and encoding categorical variables), GOEI encourages you to explore potential feature transformations that could improve predictive performance. such as Binning and/or interaction terms.\n",
    "\n",
    "**Name your final preprocessing pipeline as `prep_pipe`.**\n",
    "  \n",
    "\n",
    "# Part 2: Model Selection and Evaluation\n",
    "\n",
    "With your data preprocessed and features prepared, GOEI requests a structured approach to model selection and evaluation. Follow these steps:\n",
    "\n",
    "- Baseline Model: Start by fitting a parametric linear model as a baseline. Evaluate this model using both R² and Mean Squared Error (MSE) as performance metrics. These metrics will serve as a reference to assess improvements with other models.\n",
    "\n",
    "- Ensemble Models: GOEI is interested in seeing how ensemble methods perform on this dataset. Choose two ensemble models. To tune these models, use `RandomizedSearchCV` with 5-fold cross-validation (`cv=5`) and explore **two different hyperparameters for each model**\n",
    "\n",
    "- Performance Evaluation: For each model, evaluate performance on R² and MSE. Based on the results, choose the best-performing model and save it as `final_model`. Record the best hyperparameters used in this model as `Best_parameters`.\n",
    "\n",
    "- Test Set Evaluation: Use the `final_model` to predict on the test set and assess performance using R² and MSE. Report these metrics as evidence of the model’s effectiveness on unseen data.\n",
    "\n",
    "# 📊 GOEI’s Assessment Criteria --  70 points overall\n",
    "\n",
    "Your work will be evaluated based on the following criteria:\n",
    "\n",
    "## Pipeline Structure and Implementation: 14 points\n",
    "\n",
    "Effective and organized implementation of the pipeline, covering both data preprocessing (`prep_pipe`) and modeling steps. A structured and reusable pipeline.\n",
    "\n",
    "## Data Preprocessing and Feature Engineering (Part 1): 33 points\n",
    "\n",
    "- Evidence-based data preparation steps informed by EDA insights.\n",
    "- Thoughtful and relevant feature engineering transformations and interactions that demonstrate innovation and alignment with the problem context.\n",
    "- Attention to collinearity, data transformations, and the creation of meaningful features that capture underlying data relationships.\n",
    "\n",
    "## Model Selection, Tuning, and Evaluation (Part 2): 23 points\n",
    "\n",
    "- Appropriate choice of a baseline model and ensemble models.\n",
    "- Evidence of hyperparameter tuning using `RandomizedSearchCV` with clear documentation of parameters and cross-validation.\n",
    "- Thorough performance evaluation on both the training (cross-validation) and test sets using R² and MSE, demonstrating a standard, evidence-based approach in model selection and assessment.\n",
    "\n",
    "\n",
    "## Documentation\n",
    "\n",
    "- Use concise markdown explanations (less than a line each) to explain each preprocessing and modeling step, highlighting what you’re doing and why.\n",
    "- Summarize your final choices and rationale in a short paragraph, emphasizing the evidence that informed each decision.\n",
    "\n",
    "\n",
    "\n",
    "GOEI values a well-justified approach over specific performance numbers, so focus on following the requirements and showing thoughtful, evidence-based innovation in your work. Remember to set `random_state=42` where needed for reproducibility."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "id": "60291ef8",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "      Xp     Yp     Hsig      Dir    Depth  X-Windv  Y-Windv   U10  Season  \\\n",
      "0  52.57  27.38  0.03444  353.248  35.6057   0.3128  -1.7557  1.06  Summer   \n",
      "1  52.57  27.38  0.03459  350.817  35.6057   0.4156  -1.0176  0.40  Summer   \n",
      "2  52.57  27.38  0.03479  354.399  35.6057   0.4148  -1.7209  1.26  Summer   \n",
      "3  52.57  27.38  0.03502  348.335  35.6057   0.1343  -0.3577  1.39    Fall   \n",
      "4  52.57  27.38  0.03545  138.355  35.6057   1.9610  -0.3797  1.06  Summer   \n",
      "\n",
      "   Temperature  Wind_Speed  Wave_Steepness Wind_Dir_Category  \n",
      "0    14.045279    1.783347        0.023272              West  \n",
      "1    13.727790    1.099197        0.022787              West  \n",
      "2    14.120766    1.770185        0.025072              West  \n",
      "3    14.558437    0.382081        0.022834              West  \n",
      "4    13.679845    1.997422        0.016984              East  \n",
      "(68754, 13)\n",
      "Index(['Xp', 'Yp', 'Hsig', 'Dir', 'Depth', 'X-Windv', 'Y-Windv', 'U10',\n",
      "       'Season', 'Temperature', 'Wind_Speed', 'Wave_Steepness',\n",
      "       'Wind_Dir_Category'],\n",
      "      dtype='object')\n"
     ]
    }
   ],
   "source": [
    "import pandas as pd\n",
    "import numpy as np\n",
    "import seaborn as sns\n",
    "import matplotlib.pyplot as plt\n",
    "\n",
    "df = pd.read_csv('data/Hs.csv')\n",
    "\n",
    "print(df.head())\n",
    "print(df.shape)\n",
    "print(df.columns)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "id": "1405846e",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "(68751, 13)\n"
     ]
    }
   ],
   "source": [
    "df = df.drop_duplicates()\n",
    "print(df.shape)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "id": "3e992e66",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Xp                   float64\n",
       "Yp                   float64\n",
       "Hsig                 float64\n",
       "Dir                  float64\n",
       "Depth                float64\n",
       "X-Windv              float64\n",
       "Y-Windv              float64\n",
       "U10                  float64\n",
       "Season                object\n",
       "Temperature          float64\n",
       "Wind_Speed           float64\n",
       "Wave_Steepness       float64\n",
       "Wind_Dir_Category     object\n",
       "dtype: object"
      ]
     },
     "execution_count": 3,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df.dtypes"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "id": "5734ba11",
   "metadata": {},
   "outputs": [],
   "source": [
    "from sklearn.model_selection import train_test_split\n",
    "\n",
    "x = df.drop(columns=['Hsig'])\n",
    "y = df['Hsig']\n",
    "\n",
    "X_train, X_test, y_train, y_test = train_test_split(x, y, test_size=0.3, random_state=42)\n",
    "\n",
    "# print(x.shape)\n",
    "# print(y.shape)\n",
    "# print(X_train.shape)\n",
    "# print(X_test.shape)\n",
    "# print(y_train.shape)\n",
    "# print(y_test.shape)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "id": "fe5c7519",
   "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>Xp</th>\n",
       "      <th>Yp</th>\n",
       "      <th>Dir</th>\n",
       "      <th>Depth</th>\n",
       "      <th>X-Windv</th>\n",
       "      <th>Y-Windv</th>\n",
       "      <th>U10</th>\n",
       "      <th>Temperature</th>\n",
       "      <th>Wind_Speed</th>\n",
       "      <th>Wave_Steepness</th>\n",
       "      <th>Hsig</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>count</th>\n",
       "      <td>4.812500e+04</td>\n",
       "      <td>48125.00</td>\n",
       "      <td>48125.000000</td>\n",
       "      <td>4.812500e+04</td>\n",
       "      <td>48125.000000</td>\n",
       "      <td>48111.000000</td>\n",
       "      <td>367.000000</td>\n",
       "      <td>48119.000000</td>\n",
       "      <td>48113.000000</td>\n",
       "      <td>48117.000000</td>\n",
       "      <td>48125.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>mean</th>\n",
       "      <td>5.257000e+01</td>\n",
       "      <td>27.38</td>\n",
       "      <td>220.353672</td>\n",
       "      <td>3.560570e+01</td>\n",
       "      <td>1.695938</td>\n",
       "      <td>-1.250506</td>\n",
       "      <td>3.053188</td>\n",
       "      <td>13.798608</td>\n",
       "      <td>3.884876</td>\n",
       "      <td>0.120883</td>\n",
       "      <td>0.434984</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>std</th>\n",
       "      <td>2.842200e-14</td>\n",
       "      <td>0.00</td>\n",
       "      <td>150.766707</td>\n",
       "      <td>1.421100e-14</td>\n",
       "      <td>2.879089</td>\n",
       "      <td>2.823709</td>\n",
       "      <td>2.424080</td>\n",
       "      <td>0.497437</td>\n",
       "      <td>2.371340</td>\n",
       "      <td>0.051925</td>\n",
       "      <td>0.340381</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>min</th>\n",
       "      <td>5.257000e+01</td>\n",
       "      <td>27.38</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>3.560570e+01</td>\n",
       "      <td>-9.087300</td>\n",
       "      <td>-11.874300</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>11.564120</td>\n",
       "      <td>0.035977</td>\n",
       "      <td>0.015663</td>\n",
       "      <td>0.034440</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>25%</th>\n",
       "      <td>5.257000e+01</td>\n",
       "      <td>27.38</td>\n",
       "      <td>63.358000</td>\n",
       "      <td>3.560570e+01</td>\n",
       "      <td>-0.272400</td>\n",
       "      <td>-3.090000</td>\n",
       "      <td>1.420000</td>\n",
       "      <td>13.460457</td>\n",
       "      <td>2.094586</td>\n",
       "      <td>0.083079</td>\n",
       "      <td>0.174560</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>50%</th>\n",
       "      <td>5.257000e+01</td>\n",
       "      <td>27.38</td>\n",
       "      <td>340.038000</td>\n",
       "      <td>3.560570e+01</td>\n",
       "      <td>1.597400</td>\n",
       "      <td>-1.274100</td>\n",
       "      <td>2.450000</td>\n",
       "      <td>13.802713</td>\n",
       "      <td>3.237759</td>\n",
       "      <td>0.114889</td>\n",
       "      <td>0.317370</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>75%</th>\n",
       "      <td>5.257000e+01</td>\n",
       "      <td>27.38</td>\n",
       "      <td>353.493000</td>\n",
       "      <td>3.560570e+01</td>\n",
       "      <td>3.594300</td>\n",
       "      <td>0.665050</td>\n",
       "      <td>3.935000</td>\n",
       "      <td>14.135796</td>\n",
       "      <td>5.278546</td>\n",
       "      <td>0.150860</td>\n",
       "      <td>0.608070</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>max</th>\n",
       "      <td>5.257000e+01</td>\n",
       "      <td>27.38</td>\n",
       "      <td>359.998000</td>\n",
       "      <td>3.560570e+01</td>\n",
       "      <td>12.412800</td>\n",
       "      <td>11.782200</td>\n",
       "      <td>12.290000</td>\n",
       "      <td>16.036464</td>\n",
       "      <td>13.467091</td>\n",
       "      <td>0.335863</td>\n",
       "      <td>2.271350</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                 Xp        Yp           Dir         Depth       X-Windv  \\\n",
       "count  4.812500e+04  48125.00  48125.000000  4.812500e+04  48125.000000   \n",
       "mean   5.257000e+01     27.38    220.353672  3.560570e+01      1.695938   \n",
       "std    2.842200e-14      0.00    150.766707  1.421100e-14      2.879089   \n",
       "min    5.257000e+01     27.38      0.000000  3.560570e+01     -9.087300   \n",
       "25%    5.257000e+01     27.38     63.358000  3.560570e+01     -0.272400   \n",
       "50%    5.257000e+01     27.38    340.038000  3.560570e+01      1.597400   \n",
       "75%    5.257000e+01     27.38    353.493000  3.560570e+01      3.594300   \n",
       "max    5.257000e+01     27.38    359.998000  3.560570e+01     12.412800   \n",
       "\n",
       "            Y-Windv         U10   Temperature    Wind_Speed  Wave_Steepness  \\\n",
       "count  48111.000000  367.000000  48119.000000  48113.000000    48117.000000   \n",
       "mean      -1.250506    3.053188     13.798608      3.884876        0.120883   \n",
       "std        2.823709    2.424080      0.497437      2.371340        0.051925   \n",
       "min      -11.874300    0.000000     11.564120      0.035977        0.015663   \n",
       "25%       -3.090000    1.420000     13.460457      2.094586        0.083079   \n",
       "50%       -1.274100    2.450000     13.802713      3.237759        0.114889   \n",
       "75%        0.665050    3.935000     14.135796      5.278546        0.150860   \n",
       "max       11.782200   12.290000     16.036464     13.467091        0.335863   \n",
       "\n",
       "               Hsig  \n",
       "count  48125.000000  \n",
       "mean       0.434984  \n",
       "std        0.340381  \n",
       "min        0.034440  \n",
       "25%        0.174560  \n",
       "50%        0.317370  \n",
       "75%        0.608070  \n",
       "max        2.271350  "
      ]
     },
     "execution_count": 5,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "xy_train = pd.concat([X_train, y_train], axis=1)\n",
    "xy_train.describe()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "id": "dbd67fca",
   "metadata": {},
   "outputs": [],
   "source": [
    "#sns.pairplot(xy_train, hue='Hsig', corner=True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "id": "7c725a0d",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Index(['Xp', 'Yp', 'Dir', 'Depth', 'X-Windv', 'Y-Windv', 'U10', 'Temperature',\n",
      "       'Wind_Speed', 'Wave_Steepness'],\n",
      "      dtype='object')\n",
      "Index(['Season', 'Wind_Dir_Category'], dtype='object')\n"
     ]
    }
   ],
   "source": [
    "num_cols = X_train.select_dtypes(include=[np.number]).columns\n",
    "cat_cols = X_train.select_dtypes(exclude=[np.number]).columns\n",
    "\n",
    "print(num_cols)\n",
    "print(cat_cols)\n",
    "\n",
    "X_train_num = X_train[num_cols]\n",
    "X_train_cat = X_train[cat_cols]\n",
    "X_test_num = X_test[num_cols]\n",
    "X_test_cat = X_test[cat_cols]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "id": "66d7fa36",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<Axes: >"
      ]
     },
     "execution_count": 8,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 1000x800 with 2 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "correlation = X_train_num.corr()\n",
    "plt.figure(figsize=(10, 8))\n",
    "sns.heatmap(correlation, annot=True, cmap='coolwarm')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "id": "fde02768",
   "metadata": {},
   "outputs": [],
   "source": [
    "#decided to drop Wave_Steepness as it is highly correlated to Wind_Speed (and from knowledge we know that waves depend on wind)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "id": "7460d4d2",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([[<Axes: title={'center': 'Xp'}>, <Axes: title={'center': 'Yp'}>,\n",
       "        <Axes: title={'center': 'Dir'}>],\n",
       "       [<Axes: title={'center': 'Depth'}>,\n",
       "        <Axes: title={'center': 'X-Windv'}>,\n",
       "        <Axes: title={'center': 'Y-Windv'}>],\n",
       "       [<Axes: title={'center': 'U10'}>,\n",
       "        <Axes: title={'center': 'Temperature'}>,\n",
       "        <Axes: title={'center': 'Wind_Speed'}>],\n",
       "       [<Axes: title={'center': 'Wave_Steepness'}>, <Axes: >, <Axes: >]],\n",
       "      dtype=object)"
      ]
     },
     "execution_count": 10,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 1000x1000 with 12 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "X_train_num.hist(figsize=(10,10),bins=100)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "cef8ea2a",
   "metadata": {},
   "outputs": [],
   "source": [
    "# from the distributions, standard scalar is decided to be used on all numerical features except Dir, which shows\n",
    "# (continued) sinusoidal features and thus SplineTransformer will be used for it\n",
    "# Although Wind_Speed is slightly skewed, a log transform attempt indeed skewed it in the opposite direction, thus it was not performed in the final model"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "id": "036d7453",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<bound method IndexOpsMixin.value_counts of 67190    52.57\n",
       "53616    52.57\n",
       "42844    52.57\n",
       "41054    52.57\n",
       "65293    52.57\n",
       "         ...  \n",
       "37197    52.57\n",
       "6268     52.57\n",
       "54889    52.57\n",
       "863      52.57\n",
       "15798    52.57\n",
       "Name: Xp, Length: 48125, dtype: float64>"
      ]
     },
     "execution_count": 12,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "X_train['Xp'].value_counts # to drop as the values are all 52.57"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "id": "79105c16",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<bound method IndexOpsMixin.value_counts of 67190    27.38\n",
       "53616    27.38\n",
       "42844    27.38\n",
       "41054    27.38\n",
       "65293    27.38\n",
       "         ...  \n",
       "37197    27.38\n",
       "6268     27.38\n",
       "54889    27.38\n",
       "863      27.38\n",
       "15798    27.38\n",
       "Name: Yp, Length: 48125, dtype: float64>"
      ]
     },
     "execution_count": 13,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "X_train['Yp'].value_counts # to drop as the values are all 27.38"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "id": "023b3fa2",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<bound method IndexOpsMixin.value_counts of 67190    35.6057\n",
       "53616    35.6057\n",
       "42844    35.6057\n",
       "41054    35.6057\n",
       "65293    35.6057\n",
       "          ...   \n",
       "37197    35.6057\n",
       "6268     35.6057\n",
       "54889    35.6057\n",
       "863      35.6057\n",
       "15798    35.6057\n",
       "Name: Depth, Length: 48125, dtype: float64>"
      ]
     },
     "execution_count": 14,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "X_train['Depth'].value_counts # to drop as the values are all 35.6057"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "79f022dc",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Index(['U10'], dtype='object')\n"
     ]
    }
   ],
   "source": [
    "missing = X_train.isnull().sum() / len(X_train) \n",
    "features_to_drop = missing[missing > 0.3].index\n",
    "\n",
    "print(features_to_drop) # to drop U10 as it has more than 30% NaNs (in fact most of them are)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "id": "95bc1e34",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<style>#sk-container-id-1 {color: black;}#sk-container-id-1 pre{padding: 0;}#sk-container-id-1 div.sk-toggleable {background-color: white;}#sk-container-id-1 label.sk-toggleable__label {cursor: pointer;display: block;width: 100%;margin-bottom: 0;padding: 0.3em;box-sizing: border-box;text-align: center;}#sk-container-id-1 label.sk-toggleable__label-arrow:before {content: \"▸\";float: left;margin-right: 0.25em;color: #696969;}#sk-container-id-1 label.sk-toggleable__label-arrow:hover:before {color: black;}#sk-container-id-1 div.sk-estimator:hover label.sk-toggleable__label-arrow:before {color: black;}#sk-container-id-1 div.sk-toggleable__content {max-height: 0;max-width: 0;overflow: hidden;text-align: left;background-color: #f0f8ff;}#sk-container-id-1 div.sk-toggleable__content pre {margin: 0.2em;color: black;border-radius: 0.25em;background-color: #f0f8ff;}#sk-container-id-1 input.sk-toggleable__control:checked~div.sk-toggleable__content {max-height: 200px;max-width: 100%;overflow: auto;}#sk-container-id-1 input.sk-toggleable__control:checked~label.sk-toggleable__label-arrow:before {content: \"▾\";}#sk-container-id-1 div.sk-estimator input.sk-toggleable__control:checked~label.sk-toggleable__label {background-color: #d4ebff;}#sk-container-id-1 div.sk-label input.sk-toggleable__control:checked~label.sk-toggleable__label {background-color: #d4ebff;}#sk-container-id-1 input.sk-hidden--visually {border: 0;clip: rect(1px 1px 1px 1px);clip: rect(1px, 1px, 1px, 1px);height: 1px;margin: -1px;overflow: hidden;padding: 0;position: absolute;width: 1px;}#sk-container-id-1 div.sk-estimator {font-family: monospace;background-color: #f0f8ff;border: 1px dotted black;border-radius: 0.25em;box-sizing: border-box;margin-bottom: 0.5em;}#sk-container-id-1 div.sk-estimator:hover {background-color: #d4ebff;}#sk-container-id-1 div.sk-parallel-item::after {content: \"\";width: 100%;border-bottom: 1px solid gray;flex-grow: 1;}#sk-container-id-1 div.sk-label:hover label.sk-toggleable__label {background-color: #d4ebff;}#sk-container-id-1 div.sk-serial::before {content: \"\";position: absolute;border-left: 1px solid gray;box-sizing: border-box;top: 0;bottom: 0;left: 50%;z-index: 0;}#sk-container-id-1 div.sk-serial {display: flex;flex-direction: column;align-items: center;background-color: white;padding-right: 0.2em;padding-left: 0.2em;position: relative;}#sk-container-id-1 div.sk-item {position: relative;z-index: 1;}#sk-container-id-1 div.sk-parallel {display: flex;align-items: stretch;justify-content: center;background-color: white;position: relative;}#sk-container-id-1 div.sk-item::before, #sk-container-id-1 div.sk-parallel-item::before {content: \"\";position: absolute;border-left: 1px solid gray;box-sizing: border-box;top: 0;bottom: 0;left: 50%;z-index: -1;}#sk-container-id-1 div.sk-parallel-item {display: flex;flex-direction: column;z-index: 1;position: relative;background-color: white;}#sk-container-id-1 div.sk-parallel-item:first-child::after {align-self: flex-end;width: 50%;}#sk-container-id-1 div.sk-parallel-item:last-child::after {align-self: flex-start;width: 50%;}#sk-container-id-1 div.sk-parallel-item:only-child::after {width: 0;}#sk-container-id-1 div.sk-dashed-wrapped {border: 1px dashed gray;margin: 0 0.4em 0.5em 0.4em;box-sizing: border-box;padding-bottom: 0.4em;background-color: white;}#sk-container-id-1 div.sk-label label {font-family: monospace;font-weight: bold;display: inline-block;line-height: 1.2em;}#sk-container-id-1 div.sk-label-container {text-align: center;}#sk-container-id-1 div.sk-container {/* jupyter's `normalize.less` sets `[hidden] { display: none; }` but bootstrap.min.css set `[hidden] { display: none !important; }` so we also need the `!important` here to be able to override the default hidden behavior on the sphinx rendered scikit-learn.org. See: https://github.com/scikit-learn/scikit-learn/issues/21755 */display: inline-block !important;position: relative;}#sk-container-id-1 div.sk-text-repr-fallback {display: none;}</style><div id=\"sk-container-id-1\" class=\"sk-top-container\"><div class=\"sk-text-repr-fallback\"><pre>Pipeline(steps=[(&#x27;imputer&#x27;, SimpleImputer(strategy=&#x27;median&#x27;)),\n",
       "                (&#x27;scaler&#x27;, StandardScaler())])</pre><b>In a Jupyter environment, please rerun this cell to show the HTML representation or trust the notebook. <br />On GitHub, the HTML representation is unable to render, please try loading this page with nbviewer.org.</b></div><div class=\"sk-container\" hidden><div class=\"sk-item sk-dashed-wrapped\"><div class=\"sk-label-container\"><div class=\"sk-label sk-toggleable\"><input class=\"sk-toggleable__control sk-hidden--visually\" id=\"sk-estimator-id-1\" type=\"checkbox\" ><label for=\"sk-estimator-id-1\" class=\"sk-toggleable__label sk-toggleable__label-arrow\">Pipeline</label><div class=\"sk-toggleable__content\"><pre>Pipeline(steps=[(&#x27;imputer&#x27;, SimpleImputer(strategy=&#x27;median&#x27;)),\n",
       "                (&#x27;scaler&#x27;, StandardScaler())])</pre></div></div></div><div class=\"sk-serial\"><div class=\"sk-item\"><div class=\"sk-estimator sk-toggleable\"><input class=\"sk-toggleable__control sk-hidden--visually\" id=\"sk-estimator-id-2\" type=\"checkbox\" ><label for=\"sk-estimator-id-2\" class=\"sk-toggleable__label sk-toggleable__label-arrow\">SimpleImputer</label><div class=\"sk-toggleable__content\"><pre>SimpleImputer(strategy=&#x27;median&#x27;)</pre></div></div></div><div class=\"sk-item\"><div class=\"sk-estimator sk-toggleable\"><input class=\"sk-toggleable__control sk-hidden--visually\" id=\"sk-estimator-id-3\" type=\"checkbox\" ><label for=\"sk-estimator-id-3\" class=\"sk-toggleable__label sk-toggleable__label-arrow\">StandardScaler</label><div class=\"sk-toggleable__content\"><pre>StandardScaler()</pre></div></div></div></div></div></div></div>"
      ],
      "text/plain": [
       "Pipeline(steps=[('imputer', SimpleImputer(strategy='median')),\n",
       "                ('scaler', StandardScaler())])"
      ]
     },
     "execution_count": 16,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "from sklearn.compose import ColumnTransformer\n",
    "from sklearn.pipeline import Pipeline\n",
    "from sklearn.preprocessing import StandardScaler, RobustScaler, MinMaxScaler, LabelEncoder, OneHotEncoder, FunctionTransformer, SplineTransformer\n",
    "from sklearn.impute import SimpleImputer\n",
    "\n",
    "#\n",
    "numeric_transformer = Pipeline(steps=[\n",
    "    ('imputer', SimpleImputer(strategy='median')),\n",
    "    ('scaler', StandardScaler())\n",
    "])\n",
    "\n",
    "numeric_transformer"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "id": "2b6d1d90",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<style>#sk-container-id-2 {color: black;}#sk-container-id-2 pre{padding: 0;}#sk-container-id-2 div.sk-toggleable {background-color: white;}#sk-container-id-2 label.sk-toggleable__label {cursor: pointer;display: block;width: 100%;margin-bottom: 0;padding: 0.3em;box-sizing: border-box;text-align: center;}#sk-container-id-2 label.sk-toggleable__label-arrow:before {content: \"▸\";float: left;margin-right: 0.25em;color: #696969;}#sk-container-id-2 label.sk-toggleable__label-arrow:hover:before {color: black;}#sk-container-id-2 div.sk-estimator:hover label.sk-toggleable__label-arrow:before {color: black;}#sk-container-id-2 div.sk-toggleable__content {max-height: 0;max-width: 0;overflow: hidden;text-align: left;background-color: #f0f8ff;}#sk-container-id-2 div.sk-toggleable__content pre {margin: 0.2em;color: black;border-radius: 0.25em;background-color: #f0f8ff;}#sk-container-id-2 input.sk-toggleable__control:checked~div.sk-toggleable__content {max-height: 200px;max-width: 100%;overflow: auto;}#sk-container-id-2 input.sk-toggleable__control:checked~label.sk-toggleable__label-arrow:before {content: \"▾\";}#sk-container-id-2 div.sk-estimator input.sk-toggleable__control:checked~label.sk-toggleable__label {background-color: #d4ebff;}#sk-container-id-2 div.sk-label input.sk-toggleable__control:checked~label.sk-toggleable__label {background-color: #d4ebff;}#sk-container-id-2 input.sk-hidden--visually {border: 0;clip: rect(1px 1px 1px 1px);clip: rect(1px, 1px, 1px, 1px);height: 1px;margin: -1px;overflow: hidden;padding: 0;position: absolute;width: 1px;}#sk-container-id-2 div.sk-estimator {font-family: monospace;background-color: #f0f8ff;border: 1px dotted black;border-radius: 0.25em;box-sizing: border-box;margin-bottom: 0.5em;}#sk-container-id-2 div.sk-estimator:hover {background-color: #d4ebff;}#sk-container-id-2 div.sk-parallel-item::after {content: \"\";width: 100%;border-bottom: 1px solid gray;flex-grow: 1;}#sk-container-id-2 div.sk-label:hover label.sk-toggleable__label {background-color: #d4ebff;}#sk-container-id-2 div.sk-serial::before {content: \"\";position: absolute;border-left: 1px solid gray;box-sizing: border-box;top: 0;bottom: 0;left: 50%;z-index: 0;}#sk-container-id-2 div.sk-serial {display: flex;flex-direction: column;align-items: center;background-color: white;padding-right: 0.2em;padding-left: 0.2em;position: relative;}#sk-container-id-2 div.sk-item {position: relative;z-index: 1;}#sk-container-id-2 div.sk-parallel {display: flex;align-items: stretch;justify-content: center;background-color: white;position: relative;}#sk-container-id-2 div.sk-item::before, #sk-container-id-2 div.sk-parallel-item::before {content: \"\";position: absolute;border-left: 1px solid gray;box-sizing: border-box;top: 0;bottom: 0;left: 50%;z-index: -1;}#sk-container-id-2 div.sk-parallel-item {display: flex;flex-direction: column;z-index: 1;position: relative;background-color: white;}#sk-container-id-2 div.sk-parallel-item:first-child::after {align-self: flex-end;width: 50%;}#sk-container-id-2 div.sk-parallel-item:last-child::after {align-self: flex-start;width: 50%;}#sk-container-id-2 div.sk-parallel-item:only-child::after {width: 0;}#sk-container-id-2 div.sk-dashed-wrapped {border: 1px dashed gray;margin: 0 0.4em 0.5em 0.4em;box-sizing: border-box;padding-bottom: 0.4em;background-color: white;}#sk-container-id-2 div.sk-label label {font-family: monospace;font-weight: bold;display: inline-block;line-height: 1.2em;}#sk-container-id-2 div.sk-label-container {text-align: center;}#sk-container-id-2 div.sk-container {/* jupyter's `normalize.less` sets `[hidden] { display: none; }` but bootstrap.min.css set `[hidden] { display: none !important; }` so we also need the `!important` here to be able to override the default hidden behavior on the sphinx rendered scikit-learn.org. See: https://github.com/scikit-learn/scikit-learn/issues/21755 */display: inline-block !important;position: relative;}#sk-container-id-2 div.sk-text-repr-fallback {display: none;}</style><div id=\"sk-container-id-2\" class=\"sk-top-container\"><div class=\"sk-text-repr-fallback\"><pre>Pipeline(steps=[(&#x27;imputer&#x27;, SimpleImputer(strategy=&#x27;most_frequent&#x27;)),\n",
       "                (&#x27;encoder&#x27;, OneHotEncoder(handle_unknown=&#x27;ignore&#x27;))])</pre><b>In a Jupyter environment, please rerun this cell to show the HTML representation or trust the notebook. <br />On GitHub, the HTML representation is unable to render, please try loading this page with nbviewer.org.</b></div><div class=\"sk-container\" hidden><div class=\"sk-item sk-dashed-wrapped\"><div class=\"sk-label-container\"><div class=\"sk-label sk-toggleable\"><input class=\"sk-toggleable__control sk-hidden--visually\" id=\"sk-estimator-id-4\" type=\"checkbox\" ><label for=\"sk-estimator-id-4\" class=\"sk-toggleable__label sk-toggleable__label-arrow\">Pipeline</label><div class=\"sk-toggleable__content\"><pre>Pipeline(steps=[(&#x27;imputer&#x27;, SimpleImputer(strategy=&#x27;most_frequent&#x27;)),\n",
       "                (&#x27;encoder&#x27;, OneHotEncoder(handle_unknown=&#x27;ignore&#x27;))])</pre></div></div></div><div class=\"sk-serial\"><div class=\"sk-item\"><div class=\"sk-estimator sk-toggleable\"><input class=\"sk-toggleable__control sk-hidden--visually\" id=\"sk-estimator-id-5\" type=\"checkbox\" ><label for=\"sk-estimator-id-5\" class=\"sk-toggleable__label sk-toggleable__label-arrow\">SimpleImputer</label><div class=\"sk-toggleable__content\"><pre>SimpleImputer(strategy=&#x27;most_frequent&#x27;)</pre></div></div></div><div class=\"sk-item\"><div class=\"sk-estimator sk-toggleable\"><input class=\"sk-toggleable__control sk-hidden--visually\" id=\"sk-estimator-id-6\" type=\"checkbox\" ><label for=\"sk-estimator-id-6\" class=\"sk-toggleable__label sk-toggleable__label-arrow\">OneHotEncoder</label><div class=\"sk-toggleable__content\"><pre>OneHotEncoder(handle_unknown=&#x27;ignore&#x27;)</pre></div></div></div></div></div></div></div>"
      ],
      "text/plain": [
       "Pipeline(steps=[('imputer', SimpleImputer(strategy='most_frequent')),\n",
       "                ('encoder', OneHotEncoder(handle_unknown='ignore'))])"
      ]
     },
     "execution_count": 17,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "#referenced OneHotEncoder:\n",
    "# https://scikit-learn.org/stable/modules/generated/sklearn.preprocessing.OneHotEncoder.html\n",
    "categorical_transformer = Pipeline(steps=[\n",
    "    ('imputer', SimpleImputer(strategy='most_frequent')),\n",
    "    ('encoder', OneHotEncoder(handle_unknown='ignore'))\n",
    "])\n",
    "\n",
    "categorical_transformer"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "id": "02feae2f",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<style>#sk-container-id-3 {color: black;}#sk-container-id-3 pre{padding: 0;}#sk-container-id-3 div.sk-toggleable {background-color: white;}#sk-container-id-3 label.sk-toggleable__label {cursor: pointer;display: block;width: 100%;margin-bottom: 0;padding: 0.3em;box-sizing: border-box;text-align: center;}#sk-container-id-3 label.sk-toggleable__label-arrow:before {content: \"▸\";float: left;margin-right: 0.25em;color: #696969;}#sk-container-id-3 label.sk-toggleable__label-arrow:hover:before {color: black;}#sk-container-id-3 div.sk-estimator:hover label.sk-toggleable__label-arrow:before {color: black;}#sk-container-id-3 div.sk-toggleable__content {max-height: 0;max-width: 0;overflow: hidden;text-align: left;background-color: #f0f8ff;}#sk-container-id-3 div.sk-toggleable__content pre {margin: 0.2em;color: black;border-radius: 0.25em;background-color: #f0f8ff;}#sk-container-id-3 input.sk-toggleable__control:checked~div.sk-toggleable__content {max-height: 200px;max-width: 100%;overflow: auto;}#sk-container-id-3 input.sk-toggleable__control:checked~label.sk-toggleable__label-arrow:before {content: \"▾\";}#sk-container-id-3 div.sk-estimator input.sk-toggleable__control:checked~label.sk-toggleable__label {background-color: #d4ebff;}#sk-container-id-3 div.sk-label input.sk-toggleable__control:checked~label.sk-toggleable__label {background-color: #d4ebff;}#sk-container-id-3 input.sk-hidden--visually {border: 0;clip: rect(1px 1px 1px 1px);clip: rect(1px, 1px, 1px, 1px);height: 1px;margin: -1px;overflow: hidden;padding: 0;position: absolute;width: 1px;}#sk-container-id-3 div.sk-estimator {font-family: monospace;background-color: #f0f8ff;border: 1px dotted black;border-radius: 0.25em;box-sizing: border-box;margin-bottom: 0.5em;}#sk-container-id-3 div.sk-estimator:hover {background-color: #d4ebff;}#sk-container-id-3 div.sk-parallel-item::after {content: \"\";width: 100%;border-bottom: 1px solid gray;flex-grow: 1;}#sk-container-id-3 div.sk-label:hover label.sk-toggleable__label {background-color: #d4ebff;}#sk-container-id-3 div.sk-serial::before {content: \"\";position: absolute;border-left: 1px solid gray;box-sizing: border-box;top: 0;bottom: 0;left: 50%;z-index: 0;}#sk-container-id-3 div.sk-serial {display: flex;flex-direction: column;align-items: center;background-color: white;padding-right: 0.2em;padding-left: 0.2em;position: relative;}#sk-container-id-3 div.sk-item {position: relative;z-index: 1;}#sk-container-id-3 div.sk-parallel {display: flex;align-items: stretch;justify-content: center;background-color: white;position: relative;}#sk-container-id-3 div.sk-item::before, #sk-container-id-3 div.sk-parallel-item::before {content: \"\";position: absolute;border-left: 1px solid gray;box-sizing: border-box;top: 0;bottom: 0;left: 50%;z-index: -1;}#sk-container-id-3 div.sk-parallel-item {display: flex;flex-direction: column;z-index: 1;position: relative;background-color: white;}#sk-container-id-3 div.sk-parallel-item:first-child::after {align-self: flex-end;width: 50%;}#sk-container-id-3 div.sk-parallel-item:last-child::after {align-self: flex-start;width: 50%;}#sk-container-id-3 div.sk-parallel-item:only-child::after {width: 0;}#sk-container-id-3 div.sk-dashed-wrapped {border: 1px dashed gray;margin: 0 0.4em 0.5em 0.4em;box-sizing: border-box;padding-bottom: 0.4em;background-color: white;}#sk-container-id-3 div.sk-label label {font-family: monospace;font-weight: bold;display: inline-block;line-height: 1.2em;}#sk-container-id-3 div.sk-label-container {text-align: center;}#sk-container-id-3 div.sk-container {/* jupyter's `normalize.less` sets `[hidden] { display: none; }` but bootstrap.min.css set `[hidden] { display: none !important; }` so we also need the `!important` here to be able to override the default hidden behavior on the sphinx rendered scikit-learn.org. See: https://github.com/scikit-learn/scikit-learn/issues/21755 */display: inline-block !important;position: relative;}#sk-container-id-3 div.sk-text-repr-fallback {display: none;}</style><div id=\"sk-container-id-3\" class=\"sk-top-container\"><div class=\"sk-text-repr-fallback\"><pre>Pipeline(steps=[(&#x27;imputer&#x27;, SimpleImputer(strategy=&#x27;most_frequent&#x27;)),\n",
       "                (&#x27;scaler&#x27;, SplineTransformer())])</pre><b>In a Jupyter environment, please rerun this cell to show the HTML representation or trust the notebook. <br />On GitHub, the HTML representation is unable to render, please try loading this page with nbviewer.org.</b></div><div class=\"sk-container\" hidden><div class=\"sk-item sk-dashed-wrapped\"><div class=\"sk-label-container\"><div class=\"sk-label sk-toggleable\"><input class=\"sk-toggleable__control sk-hidden--visually\" id=\"sk-estimator-id-7\" type=\"checkbox\" ><label for=\"sk-estimator-id-7\" class=\"sk-toggleable__label sk-toggleable__label-arrow\">Pipeline</label><div class=\"sk-toggleable__content\"><pre>Pipeline(steps=[(&#x27;imputer&#x27;, SimpleImputer(strategy=&#x27;most_frequent&#x27;)),\n",
       "                (&#x27;scaler&#x27;, SplineTransformer())])</pre></div></div></div><div class=\"sk-serial\"><div class=\"sk-item\"><div class=\"sk-estimator sk-toggleable\"><input class=\"sk-toggleable__control sk-hidden--visually\" id=\"sk-estimator-id-8\" type=\"checkbox\" ><label for=\"sk-estimator-id-8\" class=\"sk-toggleable__label sk-toggleable__label-arrow\">SimpleImputer</label><div class=\"sk-toggleable__content\"><pre>SimpleImputer(strategy=&#x27;most_frequent&#x27;)</pre></div></div></div><div class=\"sk-item\"><div class=\"sk-estimator sk-toggleable\"><input class=\"sk-toggleable__control sk-hidden--visually\" id=\"sk-estimator-id-9\" type=\"checkbox\" ><label for=\"sk-estimator-id-9\" class=\"sk-toggleable__label sk-toggleable__label-arrow\">SplineTransformer</label><div class=\"sk-toggleable__content\"><pre>SplineTransformer()</pre></div></div></div></div></div></div></div>"
      ],
      "text/plain": [
       "Pipeline(steps=[('imputer', SimpleImputer(strategy='most_frequent')),\n",
       "                ('scaler', SplineTransformer())])"
      ]
     },
     "execution_count": 18,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# referenced SplineTransformer: \n",
    "# https://scikit-learn.org/1.5/modules/generated/sklearn.preprocessing.SplineTransformer.html\n",
    "\n",
    "spline_transformer = Pipeline(steps=[\n",
    "    ('imputer', SimpleImputer(strategy='most_frequent')),\n",
    "    ('scaler', SplineTransformer())\n",
    "])\n",
    "\n",
    "spline_transformer"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "id": "4c671da9",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<style>#sk-container-id-4 {color: black;}#sk-container-id-4 pre{padding: 0;}#sk-container-id-4 div.sk-toggleable {background-color: white;}#sk-container-id-4 label.sk-toggleable__label {cursor: pointer;display: block;width: 100%;margin-bottom: 0;padding: 0.3em;box-sizing: border-box;text-align: center;}#sk-container-id-4 label.sk-toggleable__label-arrow:before {content: \"▸\";float: left;margin-right: 0.25em;color: #696969;}#sk-container-id-4 label.sk-toggleable__label-arrow:hover:before {color: black;}#sk-container-id-4 div.sk-estimator:hover label.sk-toggleable__label-arrow:before {color: black;}#sk-container-id-4 div.sk-toggleable__content {max-height: 0;max-width: 0;overflow: hidden;text-align: left;background-color: #f0f8ff;}#sk-container-id-4 div.sk-toggleable__content pre {margin: 0.2em;color: black;border-radius: 0.25em;background-color: #f0f8ff;}#sk-container-id-4 input.sk-toggleable__control:checked~div.sk-toggleable__content {max-height: 200px;max-width: 100%;overflow: auto;}#sk-container-id-4 input.sk-toggleable__control:checked~label.sk-toggleable__label-arrow:before {content: \"▾\";}#sk-container-id-4 div.sk-estimator input.sk-toggleable__control:checked~label.sk-toggleable__label {background-color: #d4ebff;}#sk-container-id-4 div.sk-label input.sk-toggleable__control:checked~label.sk-toggleable__label {background-color: #d4ebff;}#sk-container-id-4 input.sk-hidden--visually {border: 0;clip: rect(1px 1px 1px 1px);clip: rect(1px, 1px, 1px, 1px);height: 1px;margin: -1px;overflow: hidden;padding: 0;position: absolute;width: 1px;}#sk-container-id-4 div.sk-estimator {font-family: monospace;background-color: #f0f8ff;border: 1px dotted black;border-radius: 0.25em;box-sizing: border-box;margin-bottom: 0.5em;}#sk-container-id-4 div.sk-estimator:hover {background-color: #d4ebff;}#sk-container-id-4 div.sk-parallel-item::after {content: \"\";width: 100%;border-bottom: 1px solid gray;flex-grow: 1;}#sk-container-id-4 div.sk-label:hover label.sk-toggleable__label {background-color: #d4ebff;}#sk-container-id-4 div.sk-serial::before {content: \"\";position: absolute;border-left: 1px solid gray;box-sizing: border-box;top: 0;bottom: 0;left: 50%;z-index: 0;}#sk-container-id-4 div.sk-serial {display: flex;flex-direction: column;align-items: center;background-color: white;padding-right: 0.2em;padding-left: 0.2em;position: relative;}#sk-container-id-4 div.sk-item {position: relative;z-index: 1;}#sk-container-id-4 div.sk-parallel {display: flex;align-items: stretch;justify-content: center;background-color: white;position: relative;}#sk-container-id-4 div.sk-item::before, #sk-container-id-4 div.sk-parallel-item::before {content: \"\";position: absolute;border-left: 1px solid gray;box-sizing: border-box;top: 0;bottom: 0;left: 50%;z-index: -1;}#sk-container-id-4 div.sk-parallel-item {display: flex;flex-direction: column;z-index: 1;position: relative;background-color: white;}#sk-container-id-4 div.sk-parallel-item:first-child::after {align-self: flex-end;width: 50%;}#sk-container-id-4 div.sk-parallel-item:last-child::after {align-self: flex-start;width: 50%;}#sk-container-id-4 div.sk-parallel-item:only-child::after {width: 0;}#sk-container-id-4 div.sk-dashed-wrapped {border: 1px dashed gray;margin: 0 0.4em 0.5em 0.4em;box-sizing: border-box;padding-bottom: 0.4em;background-color: white;}#sk-container-id-4 div.sk-label label {font-family: monospace;font-weight: bold;display: inline-block;line-height: 1.2em;}#sk-container-id-4 div.sk-label-container {text-align: center;}#sk-container-id-4 div.sk-container {/* jupyter's `normalize.less` sets `[hidden] { display: none; }` but bootstrap.min.css set `[hidden] { display: none !important; }` so we also need the `!important` here to be able to override the default hidden behavior on the sphinx rendered scikit-learn.org. See: https://github.com/scikit-learn/scikit-learn/issues/21755 */display: inline-block !important;position: relative;}#sk-container-id-4 div.sk-text-repr-fallback {display: none;}</style><div id=\"sk-container-id-4\" class=\"sk-top-container\"><div class=\"sk-text-repr-fallback\"><pre>ColumnTransformer(transformers=[(&#x27;drop_cols&#x27;, &#x27;drop&#x27;,\n",
       "                                 [&#x27;Xp&#x27;, &#x27;Yp&#x27;, &#x27;Depth&#x27;, &#x27;U10&#x27;,\n",
       "                                  &#x27;Wave_Steepness&#x27;]),\n",
       "                                (&#x27;num&#x27;,\n",
       "                                 Pipeline(steps=[(&#x27;imputer&#x27;,\n",
       "                                                  SimpleImputer(strategy=&#x27;median&#x27;)),\n",
       "                                                 (&#x27;scaler&#x27;, StandardScaler())]),\n",
       "                                 Index([&#x27;Depth&#x27;, &#x27;Temperature&#x27;, &#x27;U10&#x27;, &#x27;Wave_Steepness&#x27;, &#x27;Wind_Speed&#x27;,\n",
       "       &#x27;X-Windv&#x27;, &#x27;Xp&#x27;, &#x27;Y-Windv&#x27;, &#x27;Yp&#x27;],\n",
       "      dtype=&#x27;object&#x27;)),\n",
       "                                (&#x27;cat&#x27;,\n",
       "                                 Pipeline(steps=[(&#x27;imputer&#x27;,\n",
       "                                                  SimpleImputer(strategy=&#x27;most_frequent&#x27;)),\n",
       "                                                 (&#x27;encoder&#x27;,\n",
       "                                                  OneHotEncoder(handle_unknown=&#x27;ignore&#x27;))]),\n",
       "                                 Index([&#x27;Season&#x27;, &#x27;Wind_Dir_Category&#x27;], dtype=&#x27;object&#x27;)),\n",
       "                                (&#x27;spline&#x27;,\n",
       "                                 Pipeline(steps=[(&#x27;imputer&#x27;,\n",
       "                                                  SimpleImputer(strategy=&#x27;most_frequent&#x27;)),\n",
       "                                                 (&#x27;scaler&#x27;,\n",
       "                                                  SplineTransformer())]),\n",
       "                                 [&#x27;Dir&#x27;])])</pre><b>In a Jupyter environment, please rerun this cell to show the HTML representation or trust the notebook. <br />On GitHub, the HTML representation is unable to render, please try loading this page with nbviewer.org.</b></div><div class=\"sk-container\" hidden><div class=\"sk-item sk-dashed-wrapped\"><div class=\"sk-label-container\"><div class=\"sk-label sk-toggleable\"><input class=\"sk-toggleable__control sk-hidden--visually\" id=\"sk-estimator-id-10\" type=\"checkbox\" ><label for=\"sk-estimator-id-10\" class=\"sk-toggleable__label sk-toggleable__label-arrow\">ColumnTransformer</label><div class=\"sk-toggleable__content\"><pre>ColumnTransformer(transformers=[(&#x27;drop_cols&#x27;, &#x27;drop&#x27;,\n",
       "                                 [&#x27;Xp&#x27;, &#x27;Yp&#x27;, &#x27;Depth&#x27;, &#x27;U10&#x27;,\n",
       "                                  &#x27;Wave_Steepness&#x27;]),\n",
       "                                (&#x27;num&#x27;,\n",
       "                                 Pipeline(steps=[(&#x27;imputer&#x27;,\n",
       "                                                  SimpleImputer(strategy=&#x27;median&#x27;)),\n",
       "                                                 (&#x27;scaler&#x27;, StandardScaler())]),\n",
       "                                 Index([&#x27;Depth&#x27;, &#x27;Temperature&#x27;, &#x27;U10&#x27;, &#x27;Wave_Steepness&#x27;, &#x27;Wind_Speed&#x27;,\n",
       "       &#x27;X-Windv&#x27;, &#x27;Xp&#x27;, &#x27;Y-Windv&#x27;, &#x27;Yp&#x27;],\n",
       "      dtype=&#x27;object&#x27;)),\n",
       "                                (&#x27;cat&#x27;,\n",
       "                                 Pipeline(steps=[(&#x27;imputer&#x27;,\n",
       "                                                  SimpleImputer(strategy=&#x27;most_frequent&#x27;)),\n",
       "                                                 (&#x27;encoder&#x27;,\n",
       "                                                  OneHotEncoder(handle_unknown=&#x27;ignore&#x27;))]),\n",
       "                                 Index([&#x27;Season&#x27;, &#x27;Wind_Dir_Category&#x27;], dtype=&#x27;object&#x27;)),\n",
       "                                (&#x27;spline&#x27;,\n",
       "                                 Pipeline(steps=[(&#x27;imputer&#x27;,\n",
       "                                                  SimpleImputer(strategy=&#x27;most_frequent&#x27;)),\n",
       "                                                 (&#x27;scaler&#x27;,\n",
       "                                                  SplineTransformer())]),\n",
       "                                 [&#x27;Dir&#x27;])])</pre></div></div></div><div class=\"sk-parallel\"><div class=\"sk-parallel-item\"><div class=\"sk-item\"><div class=\"sk-label-container\"><div class=\"sk-label sk-toggleable\"><input class=\"sk-toggleable__control sk-hidden--visually\" id=\"sk-estimator-id-11\" type=\"checkbox\" ><label for=\"sk-estimator-id-11\" class=\"sk-toggleable__label sk-toggleable__label-arrow\">drop_cols</label><div class=\"sk-toggleable__content\"><pre>[&#x27;Xp&#x27;, &#x27;Yp&#x27;, &#x27;Depth&#x27;, &#x27;U10&#x27;, &#x27;Wave_Steepness&#x27;]</pre></div></div></div><div class=\"sk-serial\"><div class=\"sk-item\"><div class=\"sk-estimator sk-toggleable\"><input class=\"sk-toggleable__control sk-hidden--visually\" id=\"sk-estimator-id-12\" type=\"checkbox\" ><label for=\"sk-estimator-id-12\" class=\"sk-toggleable__label sk-toggleable__label-arrow\">drop</label><div class=\"sk-toggleable__content\"><pre>drop</pre></div></div></div></div></div></div><div class=\"sk-parallel-item\"><div class=\"sk-item\"><div class=\"sk-label-container\"><div class=\"sk-label sk-toggleable\"><input class=\"sk-toggleable__control sk-hidden--visually\" id=\"sk-estimator-id-13\" type=\"checkbox\" ><label for=\"sk-estimator-id-13\" class=\"sk-toggleable__label sk-toggleable__label-arrow\">num</label><div class=\"sk-toggleable__content\"><pre>Index([&#x27;Depth&#x27;, &#x27;Temperature&#x27;, &#x27;U10&#x27;, &#x27;Wave_Steepness&#x27;, &#x27;Wind_Speed&#x27;,\n",
       "       &#x27;X-Windv&#x27;, &#x27;Xp&#x27;, &#x27;Y-Windv&#x27;, &#x27;Yp&#x27;],\n",
       "      dtype=&#x27;object&#x27;)</pre></div></div></div><div class=\"sk-serial\"><div class=\"sk-item\"><div class=\"sk-serial\"><div class=\"sk-item\"><div class=\"sk-estimator sk-toggleable\"><input class=\"sk-toggleable__control sk-hidden--visually\" id=\"sk-estimator-id-14\" type=\"checkbox\" ><label for=\"sk-estimator-id-14\" class=\"sk-toggleable__label sk-toggleable__label-arrow\">SimpleImputer</label><div class=\"sk-toggleable__content\"><pre>SimpleImputer(strategy=&#x27;median&#x27;)</pre></div></div></div><div class=\"sk-item\"><div class=\"sk-estimator sk-toggleable\"><input class=\"sk-toggleable__control sk-hidden--visually\" id=\"sk-estimator-id-15\" type=\"checkbox\" ><label for=\"sk-estimator-id-15\" class=\"sk-toggleable__label sk-toggleable__label-arrow\">StandardScaler</label><div class=\"sk-toggleable__content\"><pre>StandardScaler()</pre></div></div></div></div></div></div></div></div><div class=\"sk-parallel-item\"><div class=\"sk-item\"><div class=\"sk-label-container\"><div class=\"sk-label sk-toggleable\"><input class=\"sk-toggleable__control sk-hidden--visually\" id=\"sk-estimator-id-16\" type=\"checkbox\" ><label for=\"sk-estimator-id-16\" class=\"sk-toggleable__label sk-toggleable__label-arrow\">cat</label><div class=\"sk-toggleable__content\"><pre>Index([&#x27;Season&#x27;, &#x27;Wind_Dir_Category&#x27;], dtype=&#x27;object&#x27;)</pre></div></div></div><div class=\"sk-serial\"><div class=\"sk-item\"><div class=\"sk-serial\"><div class=\"sk-item\"><div class=\"sk-estimator sk-toggleable\"><input class=\"sk-toggleable__control sk-hidden--visually\" id=\"sk-estimator-id-17\" type=\"checkbox\" ><label for=\"sk-estimator-id-17\" class=\"sk-toggleable__label sk-toggleable__label-arrow\">SimpleImputer</label><div class=\"sk-toggleable__content\"><pre>SimpleImputer(strategy=&#x27;most_frequent&#x27;)</pre></div></div></div><div class=\"sk-item\"><div class=\"sk-estimator sk-toggleable\"><input class=\"sk-toggleable__control sk-hidden--visually\" id=\"sk-estimator-id-18\" type=\"checkbox\" ><label for=\"sk-estimator-id-18\" class=\"sk-toggleable__label sk-toggleable__label-arrow\">OneHotEncoder</label><div class=\"sk-toggleable__content\"><pre>OneHotEncoder(handle_unknown=&#x27;ignore&#x27;)</pre></div></div></div></div></div></div></div></div><div class=\"sk-parallel-item\"><div class=\"sk-item\"><div class=\"sk-label-container\"><div class=\"sk-label sk-toggleable\"><input class=\"sk-toggleable__control sk-hidden--visually\" id=\"sk-estimator-id-19\" type=\"checkbox\" ><label for=\"sk-estimator-id-19\" class=\"sk-toggleable__label sk-toggleable__label-arrow\">spline</label><div class=\"sk-toggleable__content\"><pre>[&#x27;Dir&#x27;]</pre></div></div></div><div class=\"sk-serial\"><div class=\"sk-item\"><div class=\"sk-serial\"><div class=\"sk-item\"><div class=\"sk-estimator sk-toggleable\"><input class=\"sk-toggleable__control sk-hidden--visually\" id=\"sk-estimator-id-20\" type=\"checkbox\" ><label for=\"sk-estimator-id-20\" class=\"sk-toggleable__label sk-toggleable__label-arrow\">SimpleImputer</label><div class=\"sk-toggleable__content\"><pre>SimpleImputer(strategy=&#x27;most_frequent&#x27;)</pre></div></div></div><div class=\"sk-item\"><div class=\"sk-estimator sk-toggleable\"><input class=\"sk-toggleable__control sk-hidden--visually\" id=\"sk-estimator-id-21\" type=\"checkbox\" ><label for=\"sk-estimator-id-21\" class=\"sk-toggleable__label sk-toggleable__label-arrow\">SplineTransformer</label><div class=\"sk-toggleable__content\"><pre>SplineTransformer()</pre></div></div></div></div></div></div></div></div></div></div></div></div>"
      ],
      "text/plain": [
       "ColumnTransformer(transformers=[('drop_cols', 'drop',\n",
       "                                 ['Xp', 'Yp', 'Depth', 'U10',\n",
       "                                  'Wave_Steepness']),\n",
       "                                ('num',\n",
       "                                 Pipeline(steps=[('imputer',\n",
       "                                                  SimpleImputer(strategy='median')),\n",
       "                                                 ('scaler', StandardScaler())]),\n",
       "                                 Index(['Depth', 'Temperature', 'U10', 'Wave_Steepness', 'Wind_Speed',\n",
       "       'X-Windv', 'Xp', 'Y-Windv', 'Yp'],\n",
       "      dtype='object')),\n",
       "                                ('cat',\n",
       "                                 Pipeline(steps=[('imputer',\n",
       "                                                  SimpleImputer(strategy='most_frequent')),\n",
       "                                                 ('encoder',\n",
       "                                                  OneHotEncoder(handle_unknown='ignore'))]),\n",
       "                                 Index(['Season', 'Wind_Dir_Category'], dtype='object')),\n",
       "                                ('spline',\n",
       "                                 Pipeline(steps=[('imputer',\n",
       "                                                  SimpleImputer(strategy='most_frequent')),\n",
       "                                                 ('scaler',\n",
       "                                                  SplineTransformer())]),\n",
       "                                 ['Dir'])])"
      ]
     },
     "execution_count": 19,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "prep_pipe = ColumnTransformer(\n",
    "    transformers=[\n",
    "        (\"drop_cols\", \"drop\", [\"Xp\", \"Yp\", \"Depth\", \"U10\", \"Wave_Steepness\"]),\n",
    "        ('num', numeric_transformer, num_cols.difference(['Dir'])),\n",
    "        ('cat', categorical_transformer, cat_cols),\n",
    "        ('spline', spline_transformer, [\"Dir\"])\n",
    "    ]\n",
    ")\n",
    "\n",
    "prep_pipe"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "id": "f26635ad",
   "metadata": {},
   "outputs": [],
   "source": [
    "X_train = prep_pipe.fit_transform(X_train)\n",
    "X_test = prep_pipe.transform(X_test)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 26,
   "id": "300020e2",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "logreg_mean_cv_mse: 0.013819979737839317\n",
      "logreg_mean_cv_r2: 0.8806270132068054\n",
      "logreg_r2: 0.8818264635916027\n",
      "logreg_mse: 0.013594226087667076\n"
     ]
    }
   ],
   "source": [
    "# linear regression model\n",
    "from sklearn.linear_model import LinearRegression\n",
    "from sklearn.metrics import r2_score, mean_squared_error, make_scorer\n",
    "from sklearn.model_selection import cross_val_score, RandomizedSearchCV\n",
    "\n",
    "\n",
    "logreg = LinearRegression()\n",
    "logreg.fit(X_train,y_train)\n",
    "\n",
    "mean_cv_score_mse = -(cross_val_score(logreg, X_train, y_train, scoring='neg_mean_squared_error', cv=5, n_jobs=-1)).mean()\n",
    "print(f'logreg_mean_cv_mse: {mean_cv_score_mse}')\n",
    "\n",
    "mean_cv_score_r2 = (cross_val_score(logreg, X_train, y_train, scoring='r2', cv=5, n_jobs=-1)).mean()\n",
    "print(f'logreg_mean_cv_r2: {mean_cv_score_r2}')\n",
    "\n",
    "# prediction on test\n",
    "y_pred = logreg.predict(X_test)\n",
    "\n",
    "logreg_r2 = r2_score(y_test, y_pred)\n",
    "logreg_mse = mean_squared_error(y_test, y_pred)\n",
    "\n",
    "print(f'logreg_r2: {logreg_r2}')\n",
    "print(f'logreg_mse: {logreg_mse}')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "id": "0aa1b4b6",
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/opt/anaconda3/envs/dsml/lib/python3.8/site-packages/sklearn/model_selection/_search.py:307: UserWarning: The total space of parameters 9 is smaller than n_iter=10. Running 9 iterations. For exhaustive searches, use GridSearchCV.\n",
      "  warnings.warn(\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "best_rf_cv_params: {'n_estimators': 50, 'max_depth': 50}\n",
      "best_rf_cv_r2: 0.9662159802722169\n",
      "best_rf_cv_mse: 0.003911541384421649\n",
      "rf_predict_r2: 0.9667839694976348\n",
      "rf_predict_mse: 0.00382104354416114\n"
     ]
    }
   ],
   "source": [
    "# random forest as the first ensemble model\n",
    "\n",
    "from sklearn.ensemble import RandomForestRegressor\n",
    "\n",
    "forest = RandomForestRegressor(random_state=42)\n",
    "\n",
    "param_dist = {\n",
    "    'n_estimators': [5, 10, 50],\n",
    "    'max_depth': [5, 10, 50]\n",
    "}\n",
    "\n",
    "scoring = {\n",
    "    'neg_mse': make_scorer(mean_squared_error,greater_is_better=False),\n",
    "    'r2': make_scorer(r2_score)\n",
    "}\n",
    "\n",
    "# referenced ChatGPT \"RandomizedSearchCV Multiple Scoring\" (https://chatgpt.com/share/6737657e-d7e4-8012-bfce-abe0cce4e6ca)\n",
    "# referenced Demonstration of multi-metric evaluation on cross_val_score and GridSearchCV\n",
    "# (https://scikit-learn.org/stable/auto_examples/model_selection/plot_multi_metric_evaluation.html#sphx-glr-auto-examples-model-selection-plot-multi-metric-evaluation-py)\n",
    "# perform randomized search CV to find the best hyperparameters for the random forest model\n",
    "search = RandomizedSearchCV(estimator=forest, param_distributions=param_dist, n_iter=10, cv=5,\n",
    "                            scoring=scoring, random_state=42, n_jobs=-1, refit='neg_mse', return_train_score=True)\n",
    "search.fit(X_train, y_train)\n",
    "\n",
    "best_rf_params = search.best_params_\n",
    "best_rf_mse = search.cv_results_['mean_test_neg_mse'][search.best_index_]\n",
    "best_rf_r2 = search.cv_results_['mean_test_r2'][search.best_index_]\n",
    "print(f'best_rf_cv_params: {best_rf_params}')\n",
    "print(f'best_rf_cv_r2: {best_rf_r2}')\n",
    "print(f'best_rf_cv_mse: {np.abs(best_rf_mse)}')\n",
    "\n",
    "# make prediction on the test set\n",
    "best_rf_model = search.best_estimator_\n",
    "y_pred = best_rf_model.predict(X_test)\n",
    "\n",
    "rf_r2 = r2_score(y_test, y_pred)\n",
    "rf_mse = mean_squared_error(y_test, y_pred)\n",
    "\n",
    "print(f'rf_predict_r2: {rf_r2}')\n",
    "print(f'rf_predict_mse: {rf_mse}')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "id": "dafa3234",
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/opt/anaconda3/envs/dsml/lib/python3.8/site-packages/sklearn/model_selection/_search.py:307: UserWarning: The total space of parameters 9 is smaller than n_iter=10. Running 9 iterations. For exhaustive searches, use GridSearchCV.\n",
      "  warnings.warn(\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "best_xgb_cv_params: {'n_estimators': 20, 'max_depth': 10}\n",
      "best_xgb_cv_r2: 0.9646979936524552\n",
      "best_xgb_cv_mse: 0.0040871249662027785\n",
      "xgb_predict_r2: 0.9652365241611405\n",
      "xgb_predict_mse: 0.0039990556643190005\n"
     ]
    }
   ],
   "source": [
    "from xgboost import XGBRegressor\n",
    "\n",
    "xgb = XGBRegressor()\n",
    "\n",
    "param_dist = {\n",
    "    'n_estimators': [5, 10, 20],\n",
    "    'max_depth': [5, 10, 20]\n",
    "}\n",
    "\n",
    "scoring = {\n",
    "    'neg_mse': make_scorer(mean_squared_error,greater_is_better=False),\n",
    "    'r2': make_scorer(r2_score)\n",
    "}\n",
    "\n",
    "# referenced ChatGPT \"RandomizedSearchCV Multiple Scoring\" (https://chatgpt.com/share/6737657e-d7e4-8012-bfce-abe0cce4e6ca)\n",
    "# referenced Demonstration of multi-metric evaluation on cross_val_score and GridSearchCV\n",
    "# (https://scikit-learn.org/stable/auto_examples/model_selection/plot_multi_metric_evaluation.html#sphx-glr-auto-examples-model-selection-plot-multi-metric-evaluation-py)\n",
    "# perform randomized search CV to find the best hyperparameters for the XGBoost model\n",
    "search = RandomizedSearchCV(estimator=xgb, param_distributions=param_dist, n_iter=10, cv=5,\n",
    "                            scoring=scoring, random_state=42, n_jobs=-1, refit='neg_mse', return_train_score=True)\n",
    "search.fit(X_train, y_train)\n",
    "\n",
    "best_xgb_params = search.best_params_\n",
    "#best_xgb_scores = search.best_score_\n",
    "best_xgb_mse = search.cv_results_['mean_test_neg_mse'][search.best_index_]\n",
    "best_xgb_r2 = search.cv_results_['mean_test_r2'][search.best_index_]\n",
    "#print(f'best_xgb_scores: {np.abs(best_xgb_scores)}')\n",
    "\n",
    "print(f'best_xgb_cv_params: {best_xgb_params}')\n",
    "print(f'best_xgb_cv_r2: {best_xgb_r2}')\n",
    "print(f'best_xgb_cv_mse: {np.abs(best_xgb_mse)}')\n",
    "\n",
    "# make prediction on the test set\n",
    "best_xgb_model = search.best_estimator_\n",
    "y_pred = best_xgb_model.predict(X_test)\n",
    "\n",
    "xgb_r2 = r2_score(y_test, y_pred)\n",
    "xgb_mse = mean_squared_error(y_test, y_pred)\n",
    "\n",
    "print(f'xgb_predict_r2: {xgb_r2}')\n",
    "print(f'xgb_predict_mse: {xgb_mse}')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 24,
   "id": "35ef60c5",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "r2 score using best model: 0.96678\n",
      "mse score using best model: 0.00382\n"
     ]
    }
   ],
   "source": [
    "final_model = best_rf_model\n",
    "Best_parameters = best_rf_params\n",
    "\n",
    "#final_model.fit(X_train, y_train)\n",
    "y_pred_best = final_model.predict(X_test)\n",
    "\n",
    "best_r2 = r2_score(y_test, y_pred_best)\n",
    "best_mse = mean_squared_error(y_test, y_pred_best)\n",
    "\n",
    "print(f'r2 score using best model: {best_r2.round(5)}')\n",
    "print(f'mse score using best model: {best_mse.round(5)}')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 25,
   "id": "78f83b7e",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 1000x1000 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "#visualization\n",
    "plt.figure(figsize=(10,10))\n",
    "plt.scatter(y_test, y_pred_best)\n",
    "plt.xlabel('Actual values')\n",
    "plt.ylabel('Predicted values')\n",
    "plt.title('Predicted vs Actual Significant Wave Height using Random Forest Model')\n",
    "plt.grid()\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "ec82cf38",
   "metadata": {},
   "source": [
    "## summary\n",
    "\n",
    "- decision to drop a few features was based on:\n",
    "    - repeated values (\"Xp\", \"Yp\", \"Depth\")\n",
    "    - high number of NaNs (\"U10\")\n",
    "    - high correlation (and causality) between winds and waves (\"Wave_Steepness\")\n",
    "- scaling methods (decided mostly based on the histograms):\n",
    "    - for categorical features: OneHotEncoder (provides a unique array of 0 and 1 for each category)\n",
    "    - for numerical features except 'Dir': StandardScaler, as most of them are close to normal distributions\n",
    "        (Although Wind_Speed is skewed, after attempting log transform on it, the skewness flipped to the other direction, so log transform was withheld. Also, as WindSpeed was only slightly skewed, it might not benefit from log transform)\n",
    "    - for 'Dir': it shows a rather wavy-sinusoidal like pattern. Therefore, SplineTransformer was used.\n",
    "- best model selection:\n",
    "    - a baseline model of linear regression gives an r2 score of 0.881 and mse of 0.014\n",
    "    - ensemble models of 'random forest' and 'XGBoost' were used for comparison with the baseline linear regression model\n",
    "    - RandomizedSearch CV was performed to choose the best set of hyperparameters for the models\n",
    "    - compared to the baseline model, both models see improvements, with metrics of the followings:\n",
    "        - random forest: r2 = 0.96678, mse = 0.00382\n",
    "        - XGBoost: r2 = 0.96524, mse = 0.00400\n",
    "    - Evaluation metrics show that random forest outperforms XGBoost slightly, and was therefore chosen as the best model, with the best hyperparameters used as determined by RandomizedSearch CV.\n",
    "\n",
    "- Evaluation metric results show that the models are generally well fit, when comparing metrics on CV and test set."
   ]
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "dsml",
   "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": "3.8.20"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 5
}