{
 "cells": [
  {
   "cell_type": "markdown",
   "id": "7fb27b941602401d91542211134fc71a",
   "metadata": {},
   "source": [
    "# Fitting a Linear Simulation with XGBoost\n",
    "\n",
    "This notebook uses SHAP to demonstrate how XGBoost behaves when we fit it to simulated data where the label has a linear relationship to the features."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "id": "acae54e37e7d407bbb7b55eff062a284",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2026-04-09T21:20:52.376972Z",
     "iopub.status.busy": "2026-04-09T21:20:52.376796Z",
     "iopub.status.idle": "2026-04-09T21:20:57.730665Z",
     "shell.execute_reply": "2026-04-09T21:20:57.729493Z"
    }
   },
   "outputs": [],
   "source": [
    "import numpy as np\n",
    "import pandas as pd\n",
    "import xgboost\n",
    "from sklearn.model_selection import train_test_split\n",
    "\n",
    "import shap"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "9a63283cbaf04dbcab1f6479b197f3a8",
   "metadata": {},
   "source": [
    "## Build a simulated dataset with linear labels"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "id": "8dd0d8092fe74a7c96281538738b07e2",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2026-04-09T21:20:57.743309Z",
     "iopub.status.busy": "2026-04-09T21:20:57.741900Z",
     "iopub.status.idle": "2026-04-09T21:20:57.762771Z",
     "shell.execute_reply": "2026-04-09T21:20:57.762089Z"
    }
   },
   "outputs": [],
   "source": [
    "N = 10000\n",
    "M = 10\n",
    "np.random.seed(0)\n",
    "X_raw = np.random.randn(N, M)\n",
    "feature_names = [f\"feature {i}\" for i in range(M)]\n",
    "X = pd.DataFrame(X_raw, columns=feature_names)\n",
    "beta = np.random.randn(M)\n",
    "y = X_raw @ beta\n",
    "\n",
    "X_train, X_test, y_train, y_test = train_test_split(X, y)\n",
    "X_strain, X_valid, y_strain, y_valid = train_test_split(X_train, y_train)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "72eea5119410473aa328ad9291626812",
   "metadata": {},
   "source": [
    "## Build an XGBoost regressor"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "8edb47106e1a46a883d545849b8ab81b",
   "metadata": {},
   "source": [
    "### Train a depth 1 model"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "id": "10185d26023b46108eb7d9f57d49d2b3",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2026-04-09T21:20:57.765193Z",
     "iopub.status.busy": "2026-04-09T21:20:57.765046Z",
     "iopub.status.idle": "2026-04-09T21:21:05.128500Z",
     "shell.execute_reply": "2026-04-09T21:21:05.127546Z"
    }
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[0]\tvalidation_0-rmse:2.17988\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[1000]\tvalidation_0-rmse:0.95726\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[2000]\tvalidation_0-rmse:0.60452\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[3000]\tvalidation_0-rmse:0.41705\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[4000]\tvalidation_0-rmse:0.30822\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[5000]\tvalidation_0-rmse:0.24119\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[6000]\tvalidation_0-rmse:0.19857\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[7000]\tvalidation_0-rmse:0.17118\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[8000]\tvalidation_0-rmse:0.15386\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[9000]\tvalidation_0-rmse:0.14333\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[9999]\tvalidation_0-rmse:0.13717\n"
     ]
    },
    {
     "data": {
      "text/html": [
       "<style>#sk-container-id-1 {\n",
       "  /* Definition of color scheme common for light and dark mode */\n",
       "  --sklearn-color-text: #000;\n",
       "  --sklearn-color-text-muted: #666;\n",
       "  --sklearn-color-line: gray;\n",
       "  /* Definition of color scheme for unfitted estimators */\n",
       "  --sklearn-color-unfitted-level-0: #fff5e6;\n",
       "  --sklearn-color-unfitted-level-1: #f6e4d2;\n",
       "  --sklearn-color-unfitted-level-2: #ffe0b3;\n",
       "  --sklearn-color-unfitted-level-3: chocolate;\n",
       "  /* Definition of color scheme for fitted estimators */\n",
       "  --sklearn-color-fitted-level-0: #f0f8ff;\n",
       "  --sklearn-color-fitted-level-1: #d4ebff;\n",
       "  --sklearn-color-fitted-level-2: #b3dbfd;\n",
       "  --sklearn-color-fitted-level-3: cornflowerblue;\n",
       "}\n",
       "\n",
       "#sk-container-id-1.light {\n",
       "  /* Specific color for light theme */\n",
       "  --sklearn-color-text-on-default-background: black;\n",
       "  --sklearn-color-background: white;\n",
       "  --sklearn-color-border-box: black;\n",
       "  --sklearn-color-icon: #696969;\n",
       "}\n",
       "\n",
       "#sk-container-id-1.dark {\n",
       "  --sklearn-color-text-on-default-background: white;\n",
       "  --sklearn-color-background: #111;\n",
       "  --sklearn-color-border-box: white;\n",
       "  --sklearn-color-icon: #878787;\n",
       "}\n",
       "\n",
       "#sk-container-id-1 {\n",
       "  color: var(--sklearn-color-text);\n",
       "}\n",
       "\n",
       "#sk-container-id-1 pre {\n",
       "  padding: 0;\n",
       "}\n",
       "\n",
       "#sk-container-id-1 input.sk-hidden--visually {\n",
       "  border: 0;\n",
       "  clip: rect(1px 1px 1px 1px);\n",
       "  clip: rect(1px, 1px, 1px, 1px);\n",
       "  height: 1px;\n",
       "  margin: -1px;\n",
       "  overflow: hidden;\n",
       "  padding: 0;\n",
       "  position: absolute;\n",
       "  width: 1px;\n",
       "}\n",
       "\n",
       "#sk-container-id-1 div.sk-dashed-wrapped {\n",
       "  border: 1px dashed var(--sklearn-color-line);\n",
       "  margin: 0 0.4em 0.5em 0.4em;\n",
       "  box-sizing: border-box;\n",
       "  padding-bottom: 0.4em;\n",
       "  background-color: var(--sklearn-color-background);\n",
       "}\n",
       "\n",
       "#sk-container-id-1 div.sk-container {\n",
       "  /* jupyter's `normalize.less` sets `[hidden] { display: none; }`\n",
       "     but bootstrap.min.css set `[hidden] { display: none !important; }`\n",
       "     so we also need the `!important` here to be able to override the\n",
       "     default hidden behavior on the sphinx rendered scikit-learn.org.\n",
       "     See: https://github.com/scikit-learn/scikit-learn/issues/21755 */\n",
       "  display: inline-block !important;\n",
       "  position: relative;\n",
       "}\n",
       "\n",
       "#sk-container-id-1 div.sk-text-repr-fallback {\n",
       "  display: none;\n",
       "}\n",
       "\n",
       "div.sk-parallel-item,\n",
       "div.sk-serial,\n",
       "div.sk-item {\n",
       "  /* draw centered vertical line to link estimators */\n",
       "  background-image: linear-gradient(var(--sklearn-color-text-on-default-background), var(--sklearn-color-text-on-default-background));\n",
       "  background-size: 2px 100%;\n",
       "  background-repeat: no-repeat;\n",
       "  background-position: center center;\n",
       "}\n",
       "\n",
       "/* Parallel-specific style estimator block */\n",
       "\n",
       "#sk-container-id-1 div.sk-parallel-item::after {\n",
       "  content: \"\";\n",
       "  width: 100%;\n",
       "  border-bottom: 2px solid var(--sklearn-color-text-on-default-background);\n",
       "  flex-grow: 1;\n",
       "}\n",
       "\n",
       "#sk-container-id-1 div.sk-parallel {\n",
       "  display: flex;\n",
       "  align-items: stretch;\n",
       "  justify-content: center;\n",
       "  background-color: var(--sklearn-color-background);\n",
       "  position: relative;\n",
       "}\n",
       "\n",
       "#sk-container-id-1 div.sk-parallel-item {\n",
       "  display: flex;\n",
       "  flex-direction: column;\n",
       "}\n",
       "\n",
       "#sk-container-id-1 div.sk-parallel-item:first-child::after {\n",
       "  align-self: flex-end;\n",
       "  width: 50%;\n",
       "}\n",
       "\n",
       "#sk-container-id-1 div.sk-parallel-item:last-child::after {\n",
       "  align-self: flex-start;\n",
       "  width: 50%;\n",
       "}\n",
       "\n",
       "#sk-container-id-1 div.sk-parallel-item:only-child::after {\n",
       "  width: 0;\n",
       "}\n",
       "\n",
       "/* Serial-specific style estimator block */\n",
       "\n",
       "#sk-container-id-1 div.sk-serial {\n",
       "  display: flex;\n",
       "  flex-direction: column;\n",
       "  align-items: center;\n",
       "  background-color: var(--sklearn-color-background);\n",
       "  padding-right: 1em;\n",
       "  padding-left: 1em;\n",
       "}\n",
       "\n",
       "\n",
       "/* Toggleable style: style used for estimator/Pipeline/ColumnTransformer box that is\n",
       "clickable and can be expanded/collapsed.\n",
       "- Pipeline and ColumnTransformer use this feature and define the default style\n",
       "- Estimators will overwrite some part of the style using the `sk-estimator` class\n",
       "*/\n",
       "\n",
       "/* Pipeline and ColumnTransformer style (default) */\n",
       "\n",
       "#sk-container-id-1 div.sk-toggleable {\n",
       "  /* Default theme specific background. It is overwritten whether we have a\n",
       "  specific estimator or a Pipeline/ColumnTransformer */\n",
       "  background-color: var(--sklearn-color-background);\n",
       "}\n",
       "\n",
       "/* Toggleable label */\n",
       "#sk-container-id-1 label.sk-toggleable__label {\n",
       "  cursor: pointer;\n",
       "  display: flex;\n",
       "  width: 100%;\n",
       "  margin-bottom: 0;\n",
       "  padding: 0.5em;\n",
       "  box-sizing: border-box;\n",
       "  text-align: center;\n",
       "  align-items: center;\n",
       "  justify-content: center;\n",
       "  gap: 0.5em;\n",
       "}\n",
       "\n",
       "#sk-container-id-1 label.sk-toggleable__label .caption {\n",
       "  font-size: 0.6rem;\n",
       "  font-weight: lighter;\n",
       "  color: var(--sklearn-color-text-muted);\n",
       "}\n",
       "\n",
       "#sk-container-id-1 label.sk-toggleable__label-arrow:before {\n",
       "  /* Arrow on the left of the label */\n",
       "  content: \"▸\";\n",
       "  float: left;\n",
       "  margin-right: 0.25em;\n",
       "  color: var(--sklearn-color-icon);\n",
       "}\n",
       "\n",
       "#sk-container-id-1 label.sk-toggleable__label-arrow:hover:before {\n",
       "  color: var(--sklearn-color-text);\n",
       "}\n",
       "\n",
       "/* Toggleable content - dropdown */\n",
       "\n",
       "#sk-container-id-1 div.sk-toggleable__content {\n",
       "  display: none;\n",
       "  text-align: left;\n",
       "  /* unfitted */\n",
       "  background-color: var(--sklearn-color-unfitted-level-0);\n",
       "}\n",
       "\n",
       "#sk-container-id-1 div.sk-toggleable__content.fitted {\n",
       "  /* fitted */\n",
       "  background-color: var(--sklearn-color-fitted-level-0);\n",
       "}\n",
       "\n",
       "#sk-container-id-1 div.sk-toggleable__content pre {\n",
       "  margin: 0.2em;\n",
       "  border-radius: 0.25em;\n",
       "  color: var(--sklearn-color-text);\n",
       "  /* unfitted */\n",
       "  background-color: var(--sklearn-color-unfitted-level-0);\n",
       "}\n",
       "\n",
       "#sk-container-id-1 div.sk-toggleable__content.fitted pre {\n",
       "  /* unfitted */\n",
       "  background-color: var(--sklearn-color-fitted-level-0);\n",
       "}\n",
       "\n",
       "#sk-container-id-1 input.sk-toggleable__control:checked~div.sk-toggleable__content {\n",
       "  /* Expand drop-down */\n",
       "  display: block;\n",
       "  width: 100%;\n",
       "  overflow: visible;\n",
       "}\n",
       "\n",
       "#sk-container-id-1 input.sk-toggleable__control:checked~label.sk-toggleable__label-arrow:before {\n",
       "  content: \"▾\";\n",
       "}\n",
       "\n",
       "/* Pipeline/ColumnTransformer-specific style */\n",
       "\n",
       "#sk-container-id-1 div.sk-label input.sk-toggleable__control:checked~label.sk-toggleable__label {\n",
       "  color: var(--sklearn-color-text);\n",
       "  background-color: var(--sklearn-color-unfitted-level-2);\n",
       "}\n",
       "\n",
       "#sk-container-id-1 div.sk-label.fitted input.sk-toggleable__control:checked~label.sk-toggleable__label {\n",
       "  background-color: var(--sklearn-color-fitted-level-2);\n",
       "}\n",
       "\n",
       "/* Estimator-specific style */\n",
       "\n",
       "/* Colorize estimator box */\n",
       "#sk-container-id-1 div.sk-estimator input.sk-toggleable__control:checked~label.sk-toggleable__label {\n",
       "  /* unfitted */\n",
       "  background-color: var(--sklearn-color-unfitted-level-2);\n",
       "}\n",
       "\n",
       "#sk-container-id-1 div.sk-estimator.fitted input.sk-toggleable__control:checked~label.sk-toggleable__label {\n",
       "  /* fitted */\n",
       "  background-color: var(--sklearn-color-fitted-level-2);\n",
       "}\n",
       "\n",
       "#sk-container-id-1 div.sk-label label.sk-toggleable__label,\n",
       "#sk-container-id-1 div.sk-label label {\n",
       "  /* The background is the default theme color */\n",
       "  color: var(--sklearn-color-text-on-default-background);\n",
       "}\n",
       "\n",
       "/* On hover, darken the color of the background */\n",
       "#sk-container-id-1 div.sk-label:hover label.sk-toggleable__label {\n",
       "  color: var(--sklearn-color-text);\n",
       "  background-color: var(--sklearn-color-unfitted-level-2);\n",
       "}\n",
       "\n",
       "/* Label box, darken color on hover, fitted */\n",
       "#sk-container-id-1 div.sk-label.fitted:hover label.sk-toggleable__label.fitted {\n",
       "  color: var(--sklearn-color-text);\n",
       "  background-color: var(--sklearn-color-fitted-level-2);\n",
       "}\n",
       "\n",
       "/* Estimator label */\n",
       "\n",
       "#sk-container-id-1 div.sk-label label {\n",
       "  font-family: monospace;\n",
       "  font-weight: bold;\n",
       "  line-height: 1.2em;\n",
       "}\n",
       "\n",
       "#sk-container-id-1 div.sk-label-container {\n",
       "  text-align: center;\n",
       "}\n",
       "\n",
       "/* Estimator-specific */\n",
       "#sk-container-id-1 div.sk-estimator {\n",
       "  font-family: monospace;\n",
       "  border: 1px dotted var(--sklearn-color-border-box);\n",
       "  border-radius: 0.25em;\n",
       "  box-sizing: border-box;\n",
       "  margin-bottom: 0.5em;\n",
       "  /* unfitted */\n",
       "  background-color: var(--sklearn-color-unfitted-level-0);\n",
       "}\n",
       "\n",
       "#sk-container-id-1 div.sk-estimator.fitted {\n",
       "  /* fitted */\n",
       "  background-color: var(--sklearn-color-fitted-level-0);\n",
       "}\n",
       "\n",
       "/* on hover */\n",
       "#sk-container-id-1 div.sk-estimator:hover {\n",
       "  /* unfitted */\n",
       "  background-color: var(--sklearn-color-unfitted-level-2);\n",
       "}\n",
       "\n",
       "#sk-container-id-1 div.sk-estimator.fitted:hover {\n",
       "  /* fitted */\n",
       "  background-color: var(--sklearn-color-fitted-level-2);\n",
       "}\n",
       "\n",
       "/* Specification for estimator info (e.g. \"i\" and \"?\") */\n",
       "\n",
       "/* Common style for \"i\" and \"?\" */\n",
       "\n",
       ".sk-estimator-doc-link,\n",
       "a:link.sk-estimator-doc-link,\n",
       "a:visited.sk-estimator-doc-link {\n",
       "  float: right;\n",
       "  font-size: smaller;\n",
       "  line-height: 1em;\n",
       "  font-family: monospace;\n",
       "  background-color: var(--sklearn-color-unfitted-level-0);\n",
       "  border-radius: 1em;\n",
       "  height: 1em;\n",
       "  width: 1em;\n",
       "  text-decoration: none !important;\n",
       "  margin-left: 0.5em;\n",
       "  text-align: center;\n",
       "  /* unfitted */\n",
       "  border: var(--sklearn-color-unfitted-level-3) 1pt solid;\n",
       "  color: var(--sklearn-color-unfitted-level-3);\n",
       "}\n",
       "\n",
       ".sk-estimator-doc-link.fitted,\n",
       "a:link.sk-estimator-doc-link.fitted,\n",
       "a:visited.sk-estimator-doc-link.fitted {\n",
       "  /* fitted */\n",
       "  background-color: var(--sklearn-color-fitted-level-0);\n",
       "  border: var(--sklearn-color-fitted-level-3) 1pt solid;\n",
       "  color: var(--sklearn-color-fitted-level-3);\n",
       "}\n",
       "\n",
       "/* On hover */\n",
       "div.sk-estimator:hover .sk-estimator-doc-link:hover,\n",
       ".sk-estimator-doc-link:hover,\n",
       "div.sk-label-container:hover .sk-estimator-doc-link:hover,\n",
       ".sk-estimator-doc-link:hover {\n",
       "  /* unfitted */\n",
       "  background-color: var(--sklearn-color-unfitted-level-3);\n",
       "  border: var(--sklearn-color-fitted-level-0) 1pt solid;\n",
       "  color: var(--sklearn-color-unfitted-level-0);\n",
       "  text-decoration: none;\n",
       "}\n",
       "\n",
       "div.sk-estimator.fitted:hover .sk-estimator-doc-link.fitted:hover,\n",
       ".sk-estimator-doc-link.fitted:hover,\n",
       "div.sk-label-container:hover .sk-estimator-doc-link.fitted:hover,\n",
       ".sk-estimator-doc-link.fitted:hover {\n",
       "  /* fitted */\n",
       "  background-color: var(--sklearn-color-fitted-level-3);\n",
       "  border: var(--sklearn-color-fitted-level-0) 1pt solid;\n",
       "  color: var(--sklearn-color-fitted-level-0);\n",
       "  text-decoration: none;\n",
       "}\n",
       "\n",
       "/* Span, style for the box shown on hovering the info icon */\n",
       ".sk-estimator-doc-link span {\n",
       "  display: none;\n",
       "  z-index: 9999;\n",
       "  position: relative;\n",
       "  font-weight: normal;\n",
       "  right: .2ex;\n",
       "  padding: .5ex;\n",
       "  margin: .5ex;\n",
       "  width: min-content;\n",
       "  min-width: 20ex;\n",
       "  max-width: 50ex;\n",
       "  color: var(--sklearn-color-text);\n",
       "  box-shadow: 2pt 2pt 4pt #999;\n",
       "  /* unfitted */\n",
       "  background: var(--sklearn-color-unfitted-level-0);\n",
       "  border: .5pt solid var(--sklearn-color-unfitted-level-3);\n",
       "}\n",
       "\n",
       ".sk-estimator-doc-link.fitted span {\n",
       "  /* fitted */\n",
       "  background: var(--sklearn-color-fitted-level-0);\n",
       "  border: var(--sklearn-color-fitted-level-3);\n",
       "}\n",
       "\n",
       ".sk-estimator-doc-link:hover span {\n",
       "  display: block;\n",
       "}\n",
       "\n",
       "/* \"?\"-specific style due to the `<a>` HTML tag */\n",
       "\n",
       "#sk-container-id-1 a.estimator_doc_link {\n",
       "  float: right;\n",
       "  font-size: 1rem;\n",
       "  line-height: 1em;\n",
       "  font-family: monospace;\n",
       "  background-color: var(--sklearn-color-unfitted-level-0);\n",
       "  border-radius: 1rem;\n",
       "  height: 1rem;\n",
       "  width: 1rem;\n",
       "  text-decoration: none;\n",
       "  /* unfitted */\n",
       "  color: var(--sklearn-color-unfitted-level-1);\n",
       "  border: var(--sklearn-color-unfitted-level-1) 1pt solid;\n",
       "}\n",
       "\n",
       "#sk-container-id-1 a.estimator_doc_link.fitted {\n",
       "  /* fitted */\n",
       "  background-color: var(--sklearn-color-fitted-level-0);\n",
       "  border: var(--sklearn-color-fitted-level-1) 1pt solid;\n",
       "  color: var(--sklearn-color-fitted-level-1);\n",
       "}\n",
       "\n",
       "/* On hover */\n",
       "#sk-container-id-1 a.estimator_doc_link:hover {\n",
       "  /* unfitted */\n",
       "  background-color: var(--sklearn-color-unfitted-level-3);\n",
       "  color: var(--sklearn-color-background);\n",
       "  text-decoration: none;\n",
       "}\n",
       "\n",
       "#sk-container-id-1 a.estimator_doc_link.fitted:hover {\n",
       "  /* fitted */\n",
       "  background-color: var(--sklearn-color-fitted-level-3);\n",
       "}\n",
       "\n",
       ".estimator-table {\n",
       "    font-family: monospace;\n",
       "}\n",
       "\n",
       ".estimator-table summary {\n",
       "    padding: .5rem;\n",
       "    cursor: pointer;\n",
       "}\n",
       "\n",
       ".estimator-table summary::marker {\n",
       "    font-size: 0.7rem;\n",
       "}\n",
       "\n",
       ".estimator-table details[open] {\n",
       "    padding-left: 0.1rem;\n",
       "    padding-right: 0.1rem;\n",
       "    padding-bottom: 0.3rem;\n",
       "}\n",
       "\n",
       ".estimator-table .parameters-table {\n",
       "    margin-left: auto !important;\n",
       "    margin-right: auto !important;\n",
       "    margin-top: 0;\n",
       "}\n",
       "\n",
       ".estimator-table .parameters-table tr:nth-child(odd) {\n",
       "    background-color: #fff;\n",
       "}\n",
       "\n",
       ".estimator-table .parameters-table tr:nth-child(even) {\n",
       "    background-color: #f6f6f6;\n",
       "}\n",
       "\n",
       ".estimator-table .parameters-table tr:hover {\n",
       "    background-color: #e0e0e0;\n",
       "}\n",
       "\n",
       ".estimator-table table td {\n",
       "    border: 1px solid rgba(106, 105, 104, 0.232);\n",
       "}\n",
       "\n",
       "/*\n",
       "    `table td`is set in notebook with right text-align.\n",
       "    We need to overwrite it.\n",
       "*/\n",
       ".estimator-table table td.param {\n",
       "    text-align: left;\n",
       "    position: relative;\n",
       "    padding: 0;\n",
       "}\n",
       "\n",
       ".user-set td {\n",
       "    color:rgb(255, 94, 0);\n",
       "    text-align: left !important;\n",
       "}\n",
       "\n",
       ".user-set td.value {\n",
       "    color:rgb(255, 94, 0);\n",
       "    background-color: transparent;\n",
       "}\n",
       "\n",
       ".default td {\n",
       "    color: black;\n",
       "    text-align: left !important;\n",
       "}\n",
       "\n",
       ".user-set td i,\n",
       ".default td i {\n",
       "    color: black;\n",
       "}\n",
       "\n",
       "/*\n",
       "    Styles for parameter documentation links\n",
       "    We need styling for visited so jupyter doesn't overwrite it\n",
       "*/\n",
       "a.param-doc-link,\n",
       "a.param-doc-link:link,\n",
       "a.param-doc-link:visited {\n",
       "    text-decoration: underline dashed;\n",
       "    text-underline-offset: .3em;\n",
       "    color: inherit;\n",
       "    display: block;\n",
       "    padding: .5em;\n",
       "}\n",
       "\n",
       "/* \"hack\" to make the entire area of the cell containing the link clickable */\n",
       "a.param-doc-link::before {\n",
       "    position: absolute;\n",
       "    content: \"\";\n",
       "    inset: 0;\n",
       "}\n",
       "\n",
       ".param-doc-description {\n",
       "    display: none;\n",
       "    position: absolute;\n",
       "    z-index: 9999;\n",
       "    left: 0;\n",
       "    padding: .5ex;\n",
       "    margin-left: 1.5em;\n",
       "    color: var(--sklearn-color-text);\n",
       "    box-shadow: .3em .3em .4em #999;\n",
       "    width: max-content;\n",
       "    text-align: left;\n",
       "    max-height: 10em;\n",
       "    overflow-y: auto;\n",
       "\n",
       "    /* unfitted */\n",
       "    background: var(--sklearn-color-unfitted-level-0);\n",
       "    border: thin solid var(--sklearn-color-unfitted-level-3);\n",
       "}\n",
       "\n",
       "/* Fitted state for parameter tooltips */\n",
       ".fitted .param-doc-description {\n",
       "    /* fitted */\n",
       "    background: var(--sklearn-color-fitted-level-0);\n",
       "    border: thin solid var(--sklearn-color-fitted-level-3);\n",
       "}\n",
       "\n",
       ".param-doc-link:hover .param-doc-description {\n",
       "    display: block;\n",
       "}\n",
       "\n",
       ".copy-paste-icon {\n",
       "    background-image: url(data:image/svg+xml;base64,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);\n",
       "    background-repeat: no-repeat;\n",
       "    background-size: 14px 14px;\n",
       "    background-position: 0;\n",
       "    display: inline-block;\n",
       "    width: 14px;\n",
       "    height: 14px;\n",
       "    cursor: pointer;\n",
       "}\n",
       "</style><body><div id=\"sk-container-id-1\" class=\"sk-top-container\"><div class=\"sk-text-repr-fallback\"><pre>XGBRegressor(base_score=np.float64(0.013271975120564444), booster=None,\n",
       "             callbacks=None, colsample_bylevel=None, colsample_bynode=None,\n",
       "             colsample_bytree=None, device=None, early_stopping_rounds=20,\n",
       "             enable_categorical=False, eval_metric=None, feature_types=None,\n",
       "             feature_weights=None, gamma=None, grow_policy=None,\n",
       "             importance_type=None, interaction_constraints=None,\n",
       "             learning_rate=0.01, max_bin=None, max_cat_threshold=None,\n",
       "             max_cat_to_onehot=None, max_delta_step=None, max_depth=1,\n",
       "             max_leaves=None, min_child_weight=None, missing=nan,\n",
       "             monotone_constraints=None, multi_strategy=None, n_estimators=10000,\n",
       "             n_jobs=None, num_parallel_tree=None, ...)</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\"><div class=\"sk-estimator fitted sk-toggleable\"><input class=\"sk-toggleable__control sk-hidden--visually\" id=\"sk-estimator-id-1\" type=\"checkbox\" checked><label for=\"sk-estimator-id-1\" class=\"sk-toggleable__label fitted sk-toggleable__label-arrow\"><div><div>XGBRegressor</div></div><div><a class=\"sk-estimator-doc-link fitted\" rel=\"noreferrer\" target=\"_blank\" href=\"https://xgboost.readthedocs.io/en/release_3.2.0/python/python_api.html#xgboost.XGBRegressor\">?<span>Documentation for XGBRegressor</span></a><span class=\"sk-estimator-doc-link fitted\">i<span>Fitted</span></span></div></label><div class=\"sk-toggleable__content fitted\" data-param-prefix=\"\">\n",
       "        <div class=\"estimator-table\">\n",
       "            <details>\n",
       "                <summary>Parameters</summary>\n",
       "                <table class=\"parameters-table\">\n",
       "                  <tbody>\n",
       "                    \n",
       "        <tr class=\"default\">\n",
       "            <td><i class=\"copy-paste-icon\"\n",
       "                 onclick=\"copyToClipboard('objective',\n",
       "                          this.parentElement.nextElementSibling)\"\n",
       "            ></i></td>\n",
       "            <td class=\"param\">\n",
       "        <a class=\"param-doc-link\"\n",
       "            rel=\"noreferrer\" target=\"_blank\" href=\"https://xgboost.readthedocs.io/en/release_3.2.0/python/python_api.html#xgboost.XGBRegressor#:~:text=objective,-typing.Union%5Bstr%2C%20xgboost.sklearn._SklObjWProto%2C%20typing.Callable%5B%5Btyping.Any%2C%20typing.Any%5D%2C%20typing.Tuple%5Bnumpy.ndarray%2C%20numpy.ndarray%5D%5D%2C%20NoneType%5D\">\n",
       "            objective\n",
       "            <span class=\"param-doc-description\">objective: typing.Union[str, xgboost.sklearn._SklObjWProto, typing.Callable[[typing.Any, typing.Any], typing.Tuple[numpy.ndarray, numpy.ndarray]], NoneType]<br><br>Specify the learning task and the corresponding learning objective or a custom<br>objective function to be used.<br><br>For custom objective, see :doc:`/tutorials/custom_metric_obj` and<br>:ref:`custom-obj-metric` for more information, along with the end note for<br>function signatures.</span>\n",
       "        </a>\n",
       "    </td>\n",
       "            <td class=\"value\">&#x27;reg:squarederror&#x27;</td>\n",
       "        </tr>\n",
       "    \n",
       "\n",
       "        <tr class=\"user-set\">\n",
       "            <td><i class=\"copy-paste-icon\"\n",
       "                 onclick=\"copyToClipboard('base_score',\n",
       "                          this.parentElement.nextElementSibling)\"\n",
       "            ></i></td>\n",
       "            <td class=\"param\">\n",
       "        <a class=\"param-doc-link\"\n",
       "            rel=\"noreferrer\" target=\"_blank\" href=\"https://xgboost.readthedocs.io/en/release_3.2.0/python/python_api.html#xgboost.XGBRegressor#:~:text=base_score,-typing.Union%5Bfloat%2C%20typing.List%5Bfloat%5D%2C%20NoneType%5D\">\n",
       "            base_score\n",
       "            <span class=\"param-doc-description\">base_score: typing.Union[float, typing.List[float], NoneType]<br><br>The initial prediction score of all instances, global bias.</span>\n",
       "        </a>\n",
       "    </td>\n",
       "            <td class=\"value\">np.float64(0....1975120564444)</td>\n",
       "        </tr>\n",
       "    \n",
       "\n",
       "        <tr class=\"user-set\">\n",
       "            <td><i class=\"copy-paste-icon\"\n",
       "                 onclick=\"copyToClipboard('booster',\n",
       "                          this.parentElement.nextElementSibling)\"\n",
       "            ></i></td>\n",
       "            <td class=\"param\">booster</td>\n",
       "            <td class=\"value\">None</td>\n",
       "        </tr>\n",
       "    \n",
       "\n",
       "        <tr class=\"user-set\">\n",
       "            <td><i class=\"copy-paste-icon\"\n",
       "                 onclick=\"copyToClipboard('callbacks',\n",
       "                          this.parentElement.nextElementSibling)\"\n",
       "            ></i></td>\n",
       "            <td class=\"param\">\n",
       "        <a class=\"param-doc-link\"\n",
       "            rel=\"noreferrer\" target=\"_blank\" href=\"https://xgboost.readthedocs.io/en/release_3.2.0/python/python_api.html#xgboost.XGBRegressor#:~:text=callbacks,-typing.Optional%5Btyping.List%5Bxgboost.callback.TrainingCallback%5D%5D\">\n",
       "            callbacks\n",
       "            <span class=\"param-doc-description\">callbacks: typing.Optional[typing.List[xgboost.callback.TrainingCallback]]<br><br>List of callback functions that are applied at end of each iteration.<br>It is possible to use predefined callbacks by using<br>:ref:`Callback API <callback_api>`.<br><br>.. note::<br><br>   States in callback are not preserved during training, which means callback<br>   objects can not be reused for multiple training sessions without<br>   reinitialization or deepcopy.<br><br>.. code-block:: python<br><br>    for params in parameters_grid:<br>        # be sure to (re)initialize the callbacks before each run<br>        callbacks = [xgb.callback.LearningRateScheduler(custom_rates)]<br>        reg = xgboost.XGBRegressor(**params, callbacks=callbacks)<br>        reg.fit(X, y)</span>\n",
       "        </a>\n",
       "    </td>\n",
       "            <td class=\"value\">None</td>\n",
       "        </tr>\n",
       "    \n",
       "\n",
       "        <tr class=\"user-set\">\n",
       "            <td><i class=\"copy-paste-icon\"\n",
       "                 onclick=\"copyToClipboard('colsample_bylevel',\n",
       "                          this.parentElement.nextElementSibling)\"\n",
       "            ></i></td>\n",
       "            <td class=\"param\">\n",
       "        <a class=\"param-doc-link\"\n",
       "            rel=\"noreferrer\" target=\"_blank\" href=\"https://xgboost.readthedocs.io/en/release_3.2.0/python/python_api.html#xgboost.XGBRegressor#:~:text=colsample_bylevel,-typing.Optional%5Bfloat%5D\">\n",
       "            colsample_bylevel\n",
       "            <span class=\"param-doc-description\">colsample_bylevel: typing.Optional[float]<br><br>Subsample ratio of columns for each level.</span>\n",
       "        </a>\n",
       "    </td>\n",
       "            <td class=\"value\">None</td>\n",
       "        </tr>\n",
       "    \n",
       "\n",
       "        <tr class=\"user-set\">\n",
       "            <td><i class=\"copy-paste-icon\"\n",
       "                 onclick=\"copyToClipboard('colsample_bynode',\n",
       "                          this.parentElement.nextElementSibling)\"\n",
       "            ></i></td>\n",
       "            <td class=\"param\">\n",
       "        <a class=\"param-doc-link\"\n",
       "            rel=\"noreferrer\" target=\"_blank\" href=\"https://xgboost.readthedocs.io/en/release_3.2.0/python/python_api.html#xgboost.XGBRegressor#:~:text=colsample_bynode,-typing.Optional%5Bfloat%5D\">\n",
       "            colsample_bynode\n",
       "            <span class=\"param-doc-description\">colsample_bynode: typing.Optional[float]<br><br>Subsample ratio of columns for each split.</span>\n",
       "        </a>\n",
       "    </td>\n",
       "            <td class=\"value\">None</td>\n",
       "        </tr>\n",
       "    \n",
       "\n",
       "        <tr class=\"user-set\">\n",
       "            <td><i class=\"copy-paste-icon\"\n",
       "                 onclick=\"copyToClipboard('colsample_bytree',\n",
       "                          this.parentElement.nextElementSibling)\"\n",
       "            ></i></td>\n",
       "            <td class=\"param\">\n",
       "        <a class=\"param-doc-link\"\n",
       "            rel=\"noreferrer\" target=\"_blank\" href=\"https://xgboost.readthedocs.io/en/release_3.2.0/python/python_api.html#xgboost.XGBRegressor#:~:text=colsample_bytree,-typing.Optional%5Bfloat%5D\">\n",
       "            colsample_bytree\n",
       "            <span class=\"param-doc-description\">colsample_bytree: typing.Optional[float]<br><br>Subsample ratio of columns when constructing each tree.</span>\n",
       "        </a>\n",
       "    </td>\n",
       "            <td class=\"value\">None</td>\n",
       "        </tr>\n",
       "    \n",
       "\n",
       "        <tr class=\"user-set\">\n",
       "            <td><i class=\"copy-paste-icon\"\n",
       "                 onclick=\"copyToClipboard('device',\n",
       "                          this.parentElement.nextElementSibling)\"\n",
       "            ></i></td>\n",
       "            <td class=\"param\">\n",
       "        <a class=\"param-doc-link\"\n",
       "            rel=\"noreferrer\" target=\"_blank\" href=\"https://xgboost.readthedocs.io/en/release_3.2.0/python/python_api.html#xgboost.XGBRegressor#:~:text=device,-typing.Optional%5Bstr%5D\">\n",
       "            device\n",
       "            <span class=\"param-doc-description\">device: typing.Optional[str]<br><br>.. versionadded:: 2.0.0<br><br>Device ordinal, available options are `cpu`, `cuda`, and `gpu`.</span>\n",
       "        </a>\n",
       "    </td>\n",
       "            <td class=\"value\">None</td>\n",
       "        </tr>\n",
       "    \n",
       "\n",
       "        <tr class=\"user-set\">\n",
       "            <td><i class=\"copy-paste-icon\"\n",
       "                 onclick=\"copyToClipboard('early_stopping_rounds',\n",
       "                          this.parentElement.nextElementSibling)\"\n",
       "            ></i></td>\n",
       "            <td class=\"param\">\n",
       "        <a class=\"param-doc-link\"\n",
       "            rel=\"noreferrer\" target=\"_blank\" href=\"https://xgboost.readthedocs.io/en/release_3.2.0/python/python_api.html#xgboost.XGBRegressor#:~:text=early_stopping_rounds,-typing.Optional%5Bint%5D\">\n",
       "            early_stopping_rounds\n",
       "            <span class=\"param-doc-description\">early_stopping_rounds: typing.Optional[int]<br><br>.. versionadded:: 1.6.0<br><br>- Activates early stopping. Validation metric needs to improve at least once in<br>  every **early_stopping_rounds** round(s) to continue training.  Requires at<br>  least one item in **eval_set** in :py:meth:`fit`.<br><br>- If early stopping occurs, the model will have two additional attributes:<br>  :py:attr:`best_score` and :py:attr:`best_iteration`. These are used by the<br>  :py:meth:`predict` and :py:meth:`apply` methods to determine the optimal<br>  number of trees during inference. If users want to access the full model<br>  (including trees built after early stopping), they can specify the<br>  `iteration_range` in these inference methods. In addition, other utilities<br>  like model plotting can also use the entire model.<br><br>- If you prefer to discard the trees after `best_iteration`, consider using the<br>  callback function :py:class:`xgboost.callback.EarlyStopping`.<br><br>- If there's more than one item in **eval_set**, the last entry will be used for<br>  early stopping.  If there's more than one metric in **eval_metric**, the last<br>  metric will be used for early stopping.</span>\n",
       "        </a>\n",
       "    </td>\n",
       "            <td class=\"value\">20</td>\n",
       "        </tr>\n",
       "    \n",
       "\n",
       "        <tr class=\"user-set\">\n",
       "            <td><i class=\"copy-paste-icon\"\n",
       "                 onclick=\"copyToClipboard('enable_categorical',\n",
       "                          this.parentElement.nextElementSibling)\"\n",
       "            ></i></td>\n",
       "            <td class=\"param\">\n",
       "        <a class=\"param-doc-link\"\n",
       "            rel=\"noreferrer\" target=\"_blank\" href=\"https://xgboost.readthedocs.io/en/release_3.2.0/python/python_api.html#xgboost.XGBRegressor#:~:text=enable_categorical,-bool\">\n",
       "            enable_categorical\n",
       "            <span class=\"param-doc-description\">enable_categorical: bool<br><br>See the same parameter of :py:class:`DMatrix` for details.</span>\n",
       "        </a>\n",
       "    </td>\n",
       "            <td class=\"value\">False</td>\n",
       "        </tr>\n",
       "    \n",
       "\n",
       "        <tr class=\"user-set\">\n",
       "            <td><i class=\"copy-paste-icon\"\n",
       "                 onclick=\"copyToClipboard('eval_metric',\n",
       "                          this.parentElement.nextElementSibling)\"\n",
       "            ></i></td>\n",
       "            <td class=\"param\">\n",
       "        <a class=\"param-doc-link\"\n",
       "            rel=\"noreferrer\" target=\"_blank\" href=\"https://xgboost.readthedocs.io/en/release_3.2.0/python/python_api.html#xgboost.XGBRegressor#:~:text=eval_metric,-typing.Union%5Bstr%2C%20typing.List%5Btyping.Union%5Bstr%2C%20typing.Callable%5D%5D%2C%20typing.Callable%2C%20NoneType%5D\">\n",
       "            eval_metric\n",
       "            <span class=\"param-doc-description\">eval_metric: typing.Union[str, typing.List[typing.Union[str, typing.Callable]], typing.Callable, NoneType]<br><br>.. versionadded:: 1.6.0<br><br>Metric used for monitoring the training result and early stopping.  It can be a<br>string or list of strings as names of predefined metric in XGBoost (See<br>:doc:`/parameter`), one of the metrics in :py:mod:`sklearn.metrics`, or any<br>other user defined metric that looks like `sklearn.metrics`.<br><br>If custom objective is also provided, then custom metric should implement the<br>corresponding reverse link function.<br><br>Unlike the `scoring` parameter commonly used in scikit-learn, when a callable<br>object is provided, it's assumed to be a cost function and by default XGBoost<br>will minimize the result during early stopping.<br><br>For advanced usage on Early stopping like directly choosing to maximize instead<br>of minimize, see :py:obj:`xgboost.callback.EarlyStopping`.<br><br>See :doc:`/tutorials/custom_metric_obj` and :ref:`custom-obj-metric` for more<br>information.<br><br>.. code-block:: python<br><br>    from sklearn.datasets import load_diabetes<br>    from sklearn.metrics import mean_absolute_error<br>    X, y = load_diabetes(return_X_y=True)<br>    reg = xgb.XGBRegressor(<br>        tree_method=\"hist\",<br>        eval_metric=mean_absolute_error,<br>    )<br>    reg.fit(X, y, eval_set=[(X, y)])</span>\n",
       "        </a>\n",
       "    </td>\n",
       "            <td class=\"value\">None</td>\n",
       "        </tr>\n",
       "    \n",
       "\n",
       "        <tr class=\"user-set\">\n",
       "            <td><i class=\"copy-paste-icon\"\n",
       "                 onclick=\"copyToClipboard('feature_types',\n",
       "                          this.parentElement.nextElementSibling)\"\n",
       "            ></i></td>\n",
       "            <td class=\"param\">\n",
       "        <a class=\"param-doc-link\"\n",
       "            rel=\"noreferrer\" target=\"_blank\" href=\"https://xgboost.readthedocs.io/en/release_3.2.0/python/python_api.html#xgboost.XGBRegressor#:~:text=feature_types,-typing.Optional%5Btyping.Sequence%5Bstr%5D%5D\">\n",
       "            feature_types\n",
       "            <span class=\"param-doc-description\">feature_types: typing.Optional[typing.Sequence[str]]<br><br>.. versionadded:: 1.7.0<br><br>Used for specifying feature types without constructing a dataframe. See<br>the :py:class:`DMatrix` for details.</span>\n",
       "        </a>\n",
       "    </td>\n",
       "            <td class=\"value\">None</td>\n",
       "        </tr>\n",
       "    \n",
       "\n",
       "        <tr class=\"user-set\">\n",
       "            <td><i class=\"copy-paste-icon\"\n",
       "                 onclick=\"copyToClipboard('feature_weights',\n",
       "                          this.parentElement.nextElementSibling)\"\n",
       "            ></i></td>\n",
       "            <td class=\"param\">\n",
       "        <a class=\"param-doc-link\"\n",
       "            rel=\"noreferrer\" target=\"_blank\" href=\"https://xgboost.readthedocs.io/en/release_3.2.0/python/python_api.html#xgboost.XGBRegressor#:~:text=feature_weights,-Optional%5BArrayLike%5D\">\n",
       "            feature_weights\n",
       "            <span class=\"param-doc-description\">feature_weights: Optional[ArrayLike]<br><br>Weight for each feature, defines the probability of each feature being selected<br>when colsample is being used.  All values must be greater than 0, otherwise a<br>`ValueError` is thrown.</span>\n",
       "        </a>\n",
       "    </td>\n",
       "            <td class=\"value\">None</td>\n",
       "        </tr>\n",
       "    \n",
       "\n",
       "        <tr class=\"user-set\">\n",
       "            <td><i class=\"copy-paste-icon\"\n",
       "                 onclick=\"copyToClipboard('gamma',\n",
       "                          this.parentElement.nextElementSibling)\"\n",
       "            ></i></td>\n",
       "            <td class=\"param\">\n",
       "        <a class=\"param-doc-link\"\n",
       "            rel=\"noreferrer\" target=\"_blank\" href=\"https://xgboost.readthedocs.io/en/release_3.2.0/python/python_api.html#xgboost.XGBRegressor#:~:text=gamma,-typing.Optional%5Bfloat%5D\">\n",
       "            gamma\n",
       "            <span class=\"param-doc-description\">gamma: typing.Optional[float]<br><br>(min_split_loss) Minimum loss reduction required to make a further partition on<br>a leaf node of the tree.</span>\n",
       "        </a>\n",
       "    </td>\n",
       "            <td class=\"value\">None</td>\n",
       "        </tr>\n",
       "    \n",
       "\n",
       "        <tr class=\"user-set\">\n",
       "            <td><i class=\"copy-paste-icon\"\n",
       "                 onclick=\"copyToClipboard('grow_policy',\n",
       "                          this.parentElement.nextElementSibling)\"\n",
       "            ></i></td>\n",
       "            <td class=\"param\">\n",
       "        <a class=\"param-doc-link\"\n",
       "            rel=\"noreferrer\" target=\"_blank\" href=\"https://xgboost.readthedocs.io/en/release_3.2.0/python/python_api.html#xgboost.XGBRegressor#:~:text=grow_policy,-typing.Optional%5Bstr%5D\">\n",
       "            grow_policy\n",
       "            <span class=\"param-doc-description\">grow_policy: typing.Optional[str]<br><br>Tree growing policy.<br><br>- depthwise: Favors splitting at nodes closest to the node,<br>- lossguide: Favors splitting at nodes with highest loss change.</span>\n",
       "        </a>\n",
       "    </td>\n",
       "            <td class=\"value\">None</td>\n",
       "        </tr>\n",
       "    \n",
       "\n",
       "        <tr class=\"user-set\">\n",
       "            <td><i class=\"copy-paste-icon\"\n",
       "                 onclick=\"copyToClipboard('importance_type',\n",
       "                          this.parentElement.nextElementSibling)\"\n",
       "            ></i></td>\n",
       "            <td class=\"param\">importance_type</td>\n",
       "            <td class=\"value\">None</td>\n",
       "        </tr>\n",
       "    \n",
       "\n",
       "        <tr class=\"user-set\">\n",
       "            <td><i class=\"copy-paste-icon\"\n",
       "                 onclick=\"copyToClipboard('interaction_constraints',\n",
       "                          this.parentElement.nextElementSibling)\"\n",
       "            ></i></td>\n",
       "            <td class=\"param\">\n",
       "        <a class=\"param-doc-link\"\n",
       "            rel=\"noreferrer\" target=\"_blank\" href=\"https://xgboost.readthedocs.io/en/release_3.2.0/python/python_api.html#xgboost.XGBRegressor#:~:text=interaction_constraints,-typing.Union%5Bstr%2C%20typing.List%5Btyping.Tuple%5Bstr%5D%5D%2C%20NoneType%5D\">\n",
       "            interaction_constraints\n",
       "            <span class=\"param-doc-description\">interaction_constraints: typing.Union[str, typing.List[typing.Tuple[str]], NoneType]<br><br>Constraints for interaction representing permitted interactions.  The<br>constraints must be specified in the form of a nested list, e.g. ``[[0, 1], [2,<br>3, 4]]``, where each inner list is a group of indices of features that are<br>allowed to interact with each other.  See :doc:`tutorial<br></tutorials/feature_interaction_constraint>` for more information</span>\n",
       "        </a>\n",
       "    </td>\n",
       "            <td class=\"value\">None</td>\n",
       "        </tr>\n",
       "    \n",
       "\n",
       "        <tr class=\"user-set\">\n",
       "            <td><i class=\"copy-paste-icon\"\n",
       "                 onclick=\"copyToClipboard('learning_rate',\n",
       "                          this.parentElement.nextElementSibling)\"\n",
       "            ></i></td>\n",
       "            <td class=\"param\">\n",
       "        <a class=\"param-doc-link\"\n",
       "            rel=\"noreferrer\" target=\"_blank\" href=\"https://xgboost.readthedocs.io/en/release_3.2.0/python/python_api.html#xgboost.XGBRegressor#:~:text=learning_rate,-typing.Optional%5Bfloat%5D\">\n",
       "            learning_rate\n",
       "            <span class=\"param-doc-description\">learning_rate: typing.Optional[float]<br><br>Boosting learning rate (xgb's \"eta\")</span>\n",
       "        </a>\n",
       "    </td>\n",
       "            <td class=\"value\">0.01</td>\n",
       "        </tr>\n",
       "    \n",
       "\n",
       "        <tr class=\"user-set\">\n",
       "            <td><i class=\"copy-paste-icon\"\n",
       "                 onclick=\"copyToClipboard('max_bin',\n",
       "                          this.parentElement.nextElementSibling)\"\n",
       "            ></i></td>\n",
       "            <td class=\"param\">\n",
       "        <a class=\"param-doc-link\"\n",
       "            rel=\"noreferrer\" target=\"_blank\" href=\"https://xgboost.readthedocs.io/en/release_3.2.0/python/python_api.html#xgboost.XGBRegressor#:~:text=max_bin,-typing.Optional%5Bint%5D\">\n",
       "            max_bin\n",
       "            <span class=\"param-doc-description\">max_bin: typing.Optional[int]<br><br>If using histogram-based algorithm, maximum number of bins per feature</span>\n",
       "        </a>\n",
       "    </td>\n",
       "            <td class=\"value\">None</td>\n",
       "        </tr>\n",
       "    \n",
       "\n",
       "        <tr class=\"user-set\">\n",
       "            <td><i class=\"copy-paste-icon\"\n",
       "                 onclick=\"copyToClipboard('max_cat_threshold',\n",
       "                          this.parentElement.nextElementSibling)\"\n",
       "            ></i></td>\n",
       "            <td class=\"param\">\n",
       "        <a class=\"param-doc-link\"\n",
       "            rel=\"noreferrer\" target=\"_blank\" href=\"https://xgboost.readthedocs.io/en/release_3.2.0/python/python_api.html#xgboost.XGBRegressor#:~:text=max_cat_threshold,-typing.Optional%5Bint%5D\">\n",
       "            max_cat_threshold\n",
       "            <span class=\"param-doc-description\">max_cat_threshold: typing.Optional[int]<br><br>.. versionadded:: 1.7.0<br><br>.. note:: This parameter is experimental<br><br>Maximum number of categories considered for each split. Used only by<br>partition-based splits for preventing over-fitting. Also, `enable_categorical`<br>needs to be set to have categorical feature support. See :doc:`Categorical Data<br></tutorials/categorical>` and :ref:`cat-param` for details.</span>\n",
       "        </a>\n",
       "    </td>\n",
       "            <td class=\"value\">None</td>\n",
       "        </tr>\n",
       "    \n",
       "\n",
       "        <tr class=\"user-set\">\n",
       "            <td><i class=\"copy-paste-icon\"\n",
       "                 onclick=\"copyToClipboard('max_cat_to_onehot',\n",
       "                          this.parentElement.nextElementSibling)\"\n",
       "            ></i></td>\n",
       "            <td class=\"param\">\n",
       "        <a class=\"param-doc-link\"\n",
       "            rel=\"noreferrer\" target=\"_blank\" href=\"https://xgboost.readthedocs.io/en/release_3.2.0/python/python_api.html#xgboost.XGBRegressor#:~:text=max_cat_to_onehot,-Optional%5Bint%5D\">\n",
       "            max_cat_to_onehot\n",
       "            <span class=\"param-doc-description\">max_cat_to_onehot: Optional[int]<br><br>.. versionadded:: 1.6.0<br><br>.. note:: This parameter is experimental<br><br>A threshold for deciding whether XGBoost should use one-hot encoding based split<br>for categorical data.  When number of categories is lesser than the threshold<br>then one-hot encoding is chosen, otherwise the categories will be partitioned<br>into children nodes. Also, `enable_categorical` needs to be set to have<br>categorical feature support. See :doc:`Categorical Data<br></tutorials/categorical>` and :ref:`cat-param` for details.</span>\n",
       "        </a>\n",
       "    </td>\n",
       "            <td class=\"value\">None</td>\n",
       "        </tr>\n",
       "    \n",
       "\n",
       "        <tr class=\"user-set\">\n",
       "            <td><i class=\"copy-paste-icon\"\n",
       "                 onclick=\"copyToClipboard('max_delta_step',\n",
       "                          this.parentElement.nextElementSibling)\"\n",
       "            ></i></td>\n",
       "            <td class=\"param\">\n",
       "        <a class=\"param-doc-link\"\n",
       "            rel=\"noreferrer\" target=\"_blank\" href=\"https://xgboost.readthedocs.io/en/release_3.2.0/python/python_api.html#xgboost.XGBRegressor#:~:text=max_delta_step,-typing.Optional%5Bfloat%5D\">\n",
       "            max_delta_step\n",
       "            <span class=\"param-doc-description\">max_delta_step: typing.Optional[float]<br><br>Maximum delta step we allow each tree's weight estimation to be.</span>\n",
       "        </a>\n",
       "    </td>\n",
       "            <td class=\"value\">None</td>\n",
       "        </tr>\n",
       "    \n",
       "\n",
       "        <tr class=\"user-set\">\n",
       "            <td><i class=\"copy-paste-icon\"\n",
       "                 onclick=\"copyToClipboard('max_depth',\n",
       "                          this.parentElement.nextElementSibling)\"\n",
       "            ></i></td>\n",
       "            <td class=\"param\">\n",
       "        <a class=\"param-doc-link\"\n",
       "            rel=\"noreferrer\" target=\"_blank\" href=\"https://xgboost.readthedocs.io/en/release_3.2.0/python/python_api.html#xgboost.XGBRegressor#:~:text=max_depth,-%20typing.Optional%5Bint%5D\">\n",
       "            max_depth\n",
       "            <span class=\"param-doc-description\">max_depth:  typing.Optional[int]<br><br>Maximum tree depth for base learners.</span>\n",
       "        </a>\n",
       "    </td>\n",
       "            <td class=\"value\">1</td>\n",
       "        </tr>\n",
       "    \n",
       "\n",
       "        <tr class=\"user-set\">\n",
       "            <td><i class=\"copy-paste-icon\"\n",
       "                 onclick=\"copyToClipboard('max_leaves',\n",
       "                          this.parentElement.nextElementSibling)\"\n",
       "            ></i></td>\n",
       "            <td class=\"param\">\n",
       "        <a class=\"param-doc-link\"\n",
       "            rel=\"noreferrer\" target=\"_blank\" href=\"https://xgboost.readthedocs.io/en/release_3.2.0/python/python_api.html#xgboost.XGBRegressor#:~:text=max_leaves,-typing.Optional%5Bint%5D\">\n",
       "            max_leaves\n",
       "            <span class=\"param-doc-description\">max_leaves: typing.Optional[int]<br><br>Maximum number of leaves; 0 indicates no limit.</span>\n",
       "        </a>\n",
       "    </td>\n",
       "            <td class=\"value\">None</td>\n",
       "        </tr>\n",
       "    \n",
       "\n",
       "        <tr class=\"user-set\">\n",
       "            <td><i class=\"copy-paste-icon\"\n",
       "                 onclick=\"copyToClipboard('min_child_weight',\n",
       "                          this.parentElement.nextElementSibling)\"\n",
       "            ></i></td>\n",
       "            <td class=\"param\">\n",
       "        <a class=\"param-doc-link\"\n",
       "            rel=\"noreferrer\" target=\"_blank\" href=\"https://xgboost.readthedocs.io/en/release_3.2.0/python/python_api.html#xgboost.XGBRegressor#:~:text=min_child_weight,-typing.Optional%5Bfloat%5D\">\n",
       "            min_child_weight\n",
       "            <span class=\"param-doc-description\">min_child_weight: typing.Optional[float]<br><br>Minimum sum of instance weight(hessian) needed in a child.</span>\n",
       "        </a>\n",
       "    </td>\n",
       "            <td class=\"value\">None</td>\n",
       "        </tr>\n",
       "    \n",
       "\n",
       "        <tr class=\"user-set\">\n",
       "            <td><i class=\"copy-paste-icon\"\n",
       "                 onclick=\"copyToClipboard('missing',\n",
       "                          this.parentElement.nextElementSibling)\"\n",
       "            ></i></td>\n",
       "            <td class=\"param\">\n",
       "        <a class=\"param-doc-link\"\n",
       "            rel=\"noreferrer\" target=\"_blank\" href=\"https://xgboost.readthedocs.io/en/release_3.2.0/python/python_api.html#xgboost.XGBRegressor#:~:text=missing,-float\">\n",
       "            missing\n",
       "            <span class=\"param-doc-description\">missing: float<br><br>Value in the data which needs to be present as a missing value. Default to<br>:py:data:`numpy.nan`.</span>\n",
       "        </a>\n",
       "    </td>\n",
       "            <td class=\"value\">nan</td>\n",
       "        </tr>\n",
       "    \n",
       "\n",
       "        <tr class=\"user-set\">\n",
       "            <td><i class=\"copy-paste-icon\"\n",
       "                 onclick=\"copyToClipboard('monotone_constraints',\n",
       "                          this.parentElement.nextElementSibling)\"\n",
       "            ></i></td>\n",
       "            <td class=\"param\">\n",
       "        <a class=\"param-doc-link\"\n",
       "            rel=\"noreferrer\" target=\"_blank\" href=\"https://xgboost.readthedocs.io/en/release_3.2.0/python/python_api.html#xgboost.XGBRegressor#:~:text=monotone_constraints,-typing.Union%5Btyping.Dict%5Bstr%2C%20int%5D%2C%20str%2C%20NoneType%5D\">\n",
       "            monotone_constraints\n",
       "            <span class=\"param-doc-description\">monotone_constraints: typing.Union[typing.Dict[str, int], str, NoneType]<br><br>Constraint of variable monotonicity.  See :doc:`tutorial </tutorials/monotonic>`<br>for more information.</span>\n",
       "        </a>\n",
       "    </td>\n",
       "            <td class=\"value\">None</td>\n",
       "        </tr>\n",
       "    \n",
       "\n",
       "        <tr class=\"user-set\">\n",
       "            <td><i class=\"copy-paste-icon\"\n",
       "                 onclick=\"copyToClipboard('multi_strategy',\n",
       "                          this.parentElement.nextElementSibling)\"\n",
       "            ></i></td>\n",
       "            <td class=\"param\">\n",
       "        <a class=\"param-doc-link\"\n",
       "            rel=\"noreferrer\" target=\"_blank\" href=\"https://xgboost.readthedocs.io/en/release_3.2.0/python/python_api.html#xgboost.XGBRegressor#:~:text=multi_strategy,-typing.Optional%5Bstr%5D\">\n",
       "            multi_strategy\n",
       "            <span class=\"param-doc-description\">multi_strategy: typing.Optional[str]<br><br>.. versionadded:: 2.0.0<br><br>.. note:: This parameter is working-in-progress.<br><br>The strategy used for training multi-target models, including multi-target<br>regression and multi-class classification. See :doc:`/tutorials/multioutput` for<br>more information.<br><br>- ``one_output_per_tree``: One model for each target.<br>- ``multi_output_tree``:  Use multi-target trees.</span>\n",
       "        </a>\n",
       "    </td>\n",
       "            <td class=\"value\">None</td>\n",
       "        </tr>\n",
       "    \n",
       "\n",
       "        <tr class=\"user-set\">\n",
       "            <td><i class=\"copy-paste-icon\"\n",
       "                 onclick=\"copyToClipboard('n_estimators',\n",
       "                          this.parentElement.nextElementSibling)\"\n",
       "            ></i></td>\n",
       "            <td class=\"param\">\n",
       "        <a class=\"param-doc-link\"\n",
       "            rel=\"noreferrer\" target=\"_blank\" href=\"https://xgboost.readthedocs.io/en/release_3.2.0/python/python_api.html#xgboost.XGBRegressor#:~:text=n_estimators,-typing.Optional%5Bint%5D\">\n",
       "            n_estimators\n",
       "            <span class=\"param-doc-description\">n_estimators: typing.Optional[int]<br><br>Number of gradient boosted trees.  Equivalent to number of boosting<br>rounds.</span>\n",
       "        </a>\n",
       "    </td>\n",
       "            <td class=\"value\">10000</td>\n",
       "        </tr>\n",
       "    \n",
       "\n",
       "        <tr class=\"user-set\">\n",
       "            <td><i class=\"copy-paste-icon\"\n",
       "                 onclick=\"copyToClipboard('n_jobs',\n",
       "                          this.parentElement.nextElementSibling)\"\n",
       "            ></i></td>\n",
       "            <td class=\"param\">\n",
       "        <a class=\"param-doc-link\"\n",
       "            rel=\"noreferrer\" target=\"_blank\" href=\"https://xgboost.readthedocs.io/en/release_3.2.0/python/python_api.html#xgboost.XGBRegressor#:~:text=n_jobs,-typing.Optional%5Bint%5D\">\n",
       "            n_jobs\n",
       "            <span class=\"param-doc-description\">n_jobs: typing.Optional[int]<br><br>Number of parallel threads used to run xgboost.  When used with other<br>Scikit-Learn algorithms like grid search, you may choose which algorithm to<br>parallelize and balance the threads.  Creating thread contention will<br>significantly slow down both algorithms.</span>\n",
       "        </a>\n",
       "    </td>\n",
       "            <td class=\"value\">None</td>\n",
       "        </tr>\n",
       "    \n",
       "\n",
       "        <tr class=\"user-set\">\n",
       "            <td><i class=\"copy-paste-icon\"\n",
       "                 onclick=\"copyToClipboard('num_parallel_tree',\n",
       "                          this.parentElement.nextElementSibling)\"\n",
       "            ></i></td>\n",
       "            <td class=\"param\">num_parallel_tree</td>\n",
       "            <td class=\"value\">None</td>\n",
       "        </tr>\n",
       "    \n",
       "\n",
       "        <tr class=\"user-set\">\n",
       "            <td><i class=\"copy-paste-icon\"\n",
       "                 onclick=\"copyToClipboard('random_state',\n",
       "                          this.parentElement.nextElementSibling)\"\n",
       "            ></i></td>\n",
       "            <td class=\"param\">\n",
       "        <a class=\"param-doc-link\"\n",
       "            rel=\"noreferrer\" target=\"_blank\" href=\"https://xgboost.readthedocs.io/en/release_3.2.0/python/python_api.html#xgboost.XGBRegressor#:~:text=random_state,-typing.Union%5Bnumpy.random.mtrand.RandomState%2C%20numpy.random._generator.Generator%2C%20int%2C%20NoneType%5D\">\n",
       "            random_state\n",
       "            <span class=\"param-doc-description\">random_state: typing.Union[numpy.random.mtrand.RandomState, numpy.random._generator.Generator, int, NoneType]<br><br>Random number seed.<br><br>.. note::<br><br>   Using gblinear booster with shotgun updater is nondeterministic as<br>   it uses Hogwild algorithm.</span>\n",
       "        </a>\n",
       "    </td>\n",
       "            <td class=\"value\">None</td>\n",
       "        </tr>\n",
       "    \n",
       "\n",
       "        <tr class=\"user-set\">\n",
       "            <td><i class=\"copy-paste-icon\"\n",
       "                 onclick=\"copyToClipboard('reg_alpha',\n",
       "                          this.parentElement.nextElementSibling)\"\n",
       "            ></i></td>\n",
       "            <td class=\"param\">\n",
       "        <a class=\"param-doc-link\"\n",
       "            rel=\"noreferrer\" target=\"_blank\" href=\"https://xgboost.readthedocs.io/en/release_3.2.0/python/python_api.html#xgboost.XGBRegressor#:~:text=reg_alpha,-typing.Optional%5Bfloat%5D\">\n",
       "            reg_alpha\n",
       "            <span class=\"param-doc-description\">reg_alpha: typing.Optional[float]<br><br>L1 regularization term on weights (xgb's alpha).</span>\n",
       "        </a>\n",
       "    </td>\n",
       "            <td class=\"value\">None</td>\n",
       "        </tr>\n",
       "    \n",
       "\n",
       "        <tr class=\"user-set\">\n",
       "            <td><i class=\"copy-paste-icon\"\n",
       "                 onclick=\"copyToClipboard('reg_lambda',\n",
       "                          this.parentElement.nextElementSibling)\"\n",
       "            ></i></td>\n",
       "            <td class=\"param\">\n",
       "        <a class=\"param-doc-link\"\n",
       "            rel=\"noreferrer\" target=\"_blank\" href=\"https://xgboost.readthedocs.io/en/release_3.2.0/python/python_api.html#xgboost.XGBRegressor#:~:text=reg_lambda,-typing.Optional%5Bfloat%5D\">\n",
       "            reg_lambda\n",
       "            <span class=\"param-doc-description\">reg_lambda: typing.Optional[float]<br><br>L2 regularization term on weights (xgb's lambda).</span>\n",
       "        </a>\n",
       "    </td>\n",
       "            <td class=\"value\">None</td>\n",
       "        </tr>\n",
       "    \n",
       "\n",
       "        <tr class=\"user-set\">\n",
       "            <td><i class=\"copy-paste-icon\"\n",
       "                 onclick=\"copyToClipboard('sampling_method',\n",
       "                          this.parentElement.nextElementSibling)\"\n",
       "            ></i></td>\n",
       "            <td class=\"param\">\n",
       "        <a class=\"param-doc-link\"\n",
       "            rel=\"noreferrer\" target=\"_blank\" href=\"https://xgboost.readthedocs.io/en/release_3.2.0/python/python_api.html#xgboost.XGBRegressor#:~:text=sampling_method,-typing.Optional%5Bstr%5D\">\n",
       "            sampling_method\n",
       "            <span class=\"param-doc-description\">sampling_method: typing.Optional[str]<br><br>Sampling method. Used only by the GPU version of ``hist`` tree method.<br><br>- ``uniform``: Select random training instances uniformly.<br>- ``gradient_based``: Select random training instances with higher probability<br>    when the gradient and hessian are larger. (cf. CatBoost)</span>\n",
       "        </a>\n",
       "    </td>\n",
       "            <td class=\"value\">None</td>\n",
       "        </tr>\n",
       "    \n",
       "\n",
       "        <tr class=\"user-set\">\n",
       "            <td><i class=\"copy-paste-icon\"\n",
       "                 onclick=\"copyToClipboard('scale_pos_weight',\n",
       "                          this.parentElement.nextElementSibling)\"\n",
       "            ></i></td>\n",
       "            <td class=\"param\">\n",
       "        <a class=\"param-doc-link\"\n",
       "            rel=\"noreferrer\" target=\"_blank\" href=\"https://xgboost.readthedocs.io/en/release_3.2.0/python/python_api.html#xgboost.XGBRegressor#:~:text=scale_pos_weight,-typing.Optional%5Bfloat%5D\">\n",
       "            scale_pos_weight\n",
       "            <span class=\"param-doc-description\">scale_pos_weight: typing.Optional[float]<br><br>Balancing of positive and negative weights.</span>\n",
       "        </a>\n",
       "    </td>\n",
       "            <td class=\"value\">None</td>\n",
       "        </tr>\n",
       "    \n",
       "\n",
       "        <tr class=\"user-set\">\n",
       "            <td><i class=\"copy-paste-icon\"\n",
       "                 onclick=\"copyToClipboard('subsample',\n",
       "                          this.parentElement.nextElementSibling)\"\n",
       "            ></i></td>\n",
       "            <td class=\"param\">\n",
       "        <a class=\"param-doc-link\"\n",
       "            rel=\"noreferrer\" target=\"_blank\" href=\"https://xgboost.readthedocs.io/en/release_3.2.0/python/python_api.html#xgboost.XGBRegressor#:~:text=subsample,-typing.Optional%5Bfloat%5D\">\n",
       "            subsample\n",
       "            <span class=\"param-doc-description\">subsample: typing.Optional[float]<br><br>Subsample ratio of the training instance.</span>\n",
       "        </a>\n",
       "    </td>\n",
       "            <td class=\"value\">0.5</td>\n",
       "        </tr>\n",
       "    \n",
       "\n",
       "        <tr class=\"user-set\">\n",
       "            <td><i class=\"copy-paste-icon\"\n",
       "                 onclick=\"copyToClipboard('tree_method',\n",
       "                          this.parentElement.nextElementSibling)\"\n",
       "            ></i></td>\n",
       "            <td class=\"param\">\n",
       "        <a class=\"param-doc-link\"\n",
       "            rel=\"noreferrer\" target=\"_blank\" href=\"https://xgboost.readthedocs.io/en/release_3.2.0/python/python_api.html#xgboost.XGBRegressor#:~:text=tree_method,-typing.Optional%5Bstr%5D\">\n",
       "            tree_method\n",
       "            <span class=\"param-doc-description\">tree_method: typing.Optional[str]<br><br>Specify which tree method to use.  Default to auto.  If this parameter is set to<br>default, XGBoost will choose the most conservative option available.  It's<br>recommended to study this option from the parameters document :doc:`tree method<br></treemethod>`</span>\n",
       "        </a>\n",
       "    </td>\n",
       "            <td class=\"value\">None</td>\n",
       "        </tr>\n",
       "    \n",
       "\n",
       "        <tr class=\"user-set\">\n",
       "            <td><i class=\"copy-paste-icon\"\n",
       "                 onclick=\"copyToClipboard('validate_parameters',\n",
       "                          this.parentElement.nextElementSibling)\"\n",
       "            ></i></td>\n",
       "            <td class=\"param\">\n",
       "        <a class=\"param-doc-link\"\n",
       "            rel=\"noreferrer\" target=\"_blank\" href=\"https://xgboost.readthedocs.io/en/release_3.2.0/python/python_api.html#xgboost.XGBRegressor#:~:text=validate_parameters,-typing.Optional%5Bbool%5D\">\n",
       "            validate_parameters\n",
       "            <span class=\"param-doc-description\">validate_parameters: typing.Optional[bool]<br><br>Give warnings for unknown parameter.</span>\n",
       "        </a>\n",
       "    </td>\n",
       "            <td class=\"value\">None</td>\n",
       "        </tr>\n",
       "    \n",
       "\n",
       "        <tr class=\"user-set\">\n",
       "            <td><i class=\"copy-paste-icon\"\n",
       "                 onclick=\"copyToClipboard('verbosity',\n",
       "                          this.parentElement.nextElementSibling)\"\n",
       "            ></i></td>\n",
       "            <td class=\"param\">\n",
       "        <a class=\"param-doc-link\"\n",
       "            rel=\"noreferrer\" target=\"_blank\" href=\"https://xgboost.readthedocs.io/en/release_3.2.0/python/python_api.html#xgboost.XGBRegressor#:~:text=verbosity,-typing.Optional%5Bint%5D\">\n",
       "            verbosity\n",
       "            <span class=\"param-doc-description\">verbosity: typing.Optional[int]<br><br>The degree of verbosity. Valid values are 0 (silent) - 3 (debug).</span>\n",
       "        </a>\n",
       "    </td>\n",
       "            <td class=\"value\">None</td>\n",
       "        </tr>\n",
       "    \n",
       "                  </tbody>\n",
       "                </table>\n",
       "            </details>\n",
       "        </div>\n",
       "    </div></div></div></div></div><script>function copyToClipboard(text, element) {\n",
       "    // Get the parameter prefix from the closest toggleable content\n",
       "    const toggleableContent = element.closest('.sk-toggleable__content');\n",
       "    const paramPrefix = toggleableContent ? toggleableContent.dataset.paramPrefix : '';\n",
       "    const fullParamName = paramPrefix ? `${paramPrefix}${text}` : text;\n",
       "\n",
       "    const originalStyle = element.style;\n",
       "    const computedStyle = window.getComputedStyle(element);\n",
       "    const originalWidth = computedStyle.width;\n",
       "    const originalHTML = element.innerHTML.replace('Copied!', '');\n",
       "\n",
       "    navigator.clipboard.writeText(fullParamName)\n",
       "        .then(() => {\n",
       "            element.style.width = originalWidth;\n",
       "            element.style.color = 'green';\n",
       "            element.innerHTML = \"Copied!\";\n",
       "\n",
       "            setTimeout(() => {\n",
       "                element.innerHTML = originalHTML;\n",
       "                element.style = originalStyle;\n",
       "            }, 2000);\n",
       "        })\n",
       "        .catch(err => {\n",
       "            console.error('Failed to copy:', err);\n",
       "            element.style.color = 'red';\n",
       "            element.innerHTML = \"Failed!\";\n",
       "            setTimeout(() => {\n",
       "                element.innerHTML = originalHTML;\n",
       "                element.style = originalStyle;\n",
       "            }, 2000);\n",
       "        });\n",
       "    return false;\n",
       "}\n",
       "\n",
       "document.querySelectorAll('.copy-paste-icon').forEach(function(element) {\n",
       "    const toggleableContent = element.closest('.sk-toggleable__content');\n",
       "    const paramPrefix = toggleableContent ? toggleableContent.dataset.paramPrefix : '';\n",
       "    const paramName = element.parentElement.nextElementSibling\n",
       "        .textContent.trim().split(' ')[0];\n",
       "    const fullParamName = paramPrefix ? `${paramPrefix}${paramName}` : paramName;\n",
       "\n",
       "    element.setAttribute('title', fullParamName);\n",
       "});\n",
       "\n",
       "\n",
       "/**\n",
       " * Adapted from Skrub\n",
       " * https://github.com/skrub-data/skrub/blob/403466d1d5d4dc76a7ef569b3f8228db59a31dc3/skrub/_reporting/_data/templates/report.js#L789\n",
       " * @returns \"light\" or \"dark\"\n",
       " */\n",
       "function detectTheme(element) {\n",
       "    const body = document.querySelector('body');\n",
       "\n",
       "    // Check VSCode theme\n",
       "    const themeKindAttr = body.getAttribute('data-vscode-theme-kind');\n",
       "    const themeNameAttr = body.getAttribute('data-vscode-theme-name');\n",
       "\n",
       "    if (themeKindAttr && themeNameAttr) {\n",
       "        const themeKind = themeKindAttr.toLowerCase();\n",
       "        const themeName = themeNameAttr.toLowerCase();\n",
       "\n",
       "        if (themeKind.includes(\"dark\") || themeName.includes(\"dark\")) {\n",
       "            return \"dark\";\n",
       "        }\n",
       "        if (themeKind.includes(\"light\") || themeName.includes(\"light\")) {\n",
       "            return \"light\";\n",
       "        }\n",
       "    }\n",
       "\n",
       "    // Check Jupyter theme\n",
       "    if (body.getAttribute('data-jp-theme-light') === 'false') {\n",
       "        return 'dark';\n",
       "    } else if (body.getAttribute('data-jp-theme-light') === 'true') {\n",
       "        return 'light';\n",
       "    }\n",
       "\n",
       "    // Guess based on a parent element's color\n",
       "    const color = window.getComputedStyle(element.parentNode, null).getPropertyValue('color');\n",
       "    const match = color.match(/^rgb\\s*\\(\\s*(\\d+)\\s*,\\s*(\\d+)\\s*,\\s*(\\d+)\\s*\\)\\s*$/i);\n",
       "    if (match) {\n",
       "        const [r, g, b] = [\n",
       "            parseFloat(match[1]),\n",
       "            parseFloat(match[2]),\n",
       "            parseFloat(match[3])\n",
       "        ];\n",
       "\n",
       "        // https://en.wikipedia.org/wiki/HSL_and_HSV#Lightness\n",
       "        const luma = 0.299 * r + 0.587 * g + 0.114 * b;\n",
       "\n",
       "        if (luma > 180) {\n",
       "            // If the text is very bright we have a dark theme\n",
       "            return 'dark';\n",
       "        }\n",
       "        if (luma < 75) {\n",
       "            // If the text is very dark we have a light theme\n",
       "            return 'light';\n",
       "        }\n",
       "        // Otherwise fall back to the next heuristic.\n",
       "    }\n",
       "\n",
       "    // Fallback to system preference\n",
       "    return window.matchMedia('(prefers-color-scheme: dark)').matches ? 'dark' : 'light';\n",
       "}\n",
       "\n",
       "\n",
       "function forceTheme(elementId) {\n",
       "    const estimatorElement = document.querySelector(`#${elementId}`);\n",
       "    if (estimatorElement === null) {\n",
       "        console.error(`Element with id ${elementId} not found.`);\n",
       "    } else {\n",
       "        const theme = detectTheme(estimatorElement);\n",
       "        estimatorElement.classList.add(theme);\n",
       "    }\n",
       "}\n",
       "\n",
       "forceTheme('sk-container-id-1');</script></body>"
      ],
      "text/plain": [
       "XGBRegressor(base_score=np.float64(0.013271975120564444), booster=None,\n",
       "             callbacks=None, colsample_bylevel=None, colsample_bynode=None,\n",
       "             colsample_bytree=None, device=None, early_stopping_rounds=20,\n",
       "             enable_categorical=False, eval_metric=None, feature_types=None,\n",
       "             feature_weights=None, gamma=None, grow_policy=None,\n",
       "             importance_type=None, interaction_constraints=None,\n",
       "             learning_rate=0.01, max_bin=None, max_cat_threshold=None,\n",
       "             max_cat_to_onehot=None, max_delta_step=None, max_depth=1,\n",
       "             max_leaves=None, min_child_weight=None, missing=nan,\n",
       "             monotone_constraints=None, multi_strategy=None, n_estimators=10000,\n",
       "             n_jobs=None, num_parallel_tree=None, ...)"
      ]
     },
     "execution_count": 3,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "model_depth1 = xgboost.XGBRegressor(\n",
    "    max_depth=1,\n",
    "    learning_rate=0.01,\n",
    "    subsample=0.5,\n",
    "    n_estimators=10000,\n",
    "    base_score=y_strain.mean(),\n",
    "    early_stopping_rounds=20,\n",
    ")\n",
    "model_depth1.fit(\n",
    "    X_strain,\n",
    "    y_strain,\n",
    "    eval_set=[(X_valid, y_valid)],\n",
    "    verbose=1000,\n",
    ")"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "8763a12b2bbd4a93a75aff182afb95dc",
   "metadata": {},
   "source": [
    "### Train a depth 3 model"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "id": "7623eae2785240b9bd12b16a66d81610",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2026-04-09T21:21:05.131458Z",
     "iopub.status.busy": "2026-04-09T21:21:05.131290Z",
     "iopub.status.idle": "2026-04-09T21:21:10.070031Z",
     "shell.execute_reply": "2026-04-09T21:21:10.068944Z"
    }
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[0]\tvalidation_0-rmse:2.17743\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[500]\tvalidation_0-rmse:0.27682\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[1000]\tvalidation_0-rmse:0.22374\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[1500]\tvalidation_0-rmse:0.21615\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[1655]\tvalidation_0-rmse:0.21504\n"
     ]
    },
    {
     "data": {
      "text/html": [
       "<style>#sk-container-id-2 {\n",
       "  /* Definition of color scheme common for light and dark mode */\n",
       "  --sklearn-color-text: #000;\n",
       "  --sklearn-color-text-muted: #666;\n",
       "  --sklearn-color-line: gray;\n",
       "  /* Definition of color scheme for unfitted estimators */\n",
       "  --sklearn-color-unfitted-level-0: #fff5e6;\n",
       "  --sklearn-color-unfitted-level-1: #f6e4d2;\n",
       "  --sklearn-color-unfitted-level-2: #ffe0b3;\n",
       "  --sklearn-color-unfitted-level-3: chocolate;\n",
       "  /* Definition of color scheme for fitted estimators */\n",
       "  --sklearn-color-fitted-level-0: #f0f8ff;\n",
       "  --sklearn-color-fitted-level-1: #d4ebff;\n",
       "  --sklearn-color-fitted-level-2: #b3dbfd;\n",
       "  --sklearn-color-fitted-level-3: cornflowerblue;\n",
       "}\n",
       "\n",
       "#sk-container-id-2.light {\n",
       "  /* Specific color for light theme */\n",
       "  --sklearn-color-text-on-default-background: black;\n",
       "  --sklearn-color-background: white;\n",
       "  --sklearn-color-border-box: black;\n",
       "  --sklearn-color-icon: #696969;\n",
       "}\n",
       "\n",
       "#sk-container-id-2.dark {\n",
       "  --sklearn-color-text-on-default-background: white;\n",
       "  --sklearn-color-background: #111;\n",
       "  --sklearn-color-border-box: white;\n",
       "  --sklearn-color-icon: #878787;\n",
       "}\n",
       "\n",
       "#sk-container-id-2 {\n",
       "  color: var(--sklearn-color-text);\n",
       "}\n",
       "\n",
       "#sk-container-id-2 pre {\n",
       "  padding: 0;\n",
       "}\n",
       "\n",
       "#sk-container-id-2 input.sk-hidden--visually {\n",
       "  border: 0;\n",
       "  clip: rect(1px 1px 1px 1px);\n",
       "  clip: rect(1px, 1px, 1px, 1px);\n",
       "  height: 1px;\n",
       "  margin: -1px;\n",
       "  overflow: hidden;\n",
       "  padding: 0;\n",
       "  position: absolute;\n",
       "  width: 1px;\n",
       "}\n",
       "\n",
       "#sk-container-id-2 div.sk-dashed-wrapped {\n",
       "  border: 1px dashed var(--sklearn-color-line);\n",
       "  margin: 0 0.4em 0.5em 0.4em;\n",
       "  box-sizing: border-box;\n",
       "  padding-bottom: 0.4em;\n",
       "  background-color: var(--sklearn-color-background);\n",
       "}\n",
       "\n",
       "#sk-container-id-2 div.sk-container {\n",
       "  /* jupyter's `normalize.less` sets `[hidden] { display: none; }`\n",
       "     but bootstrap.min.css set `[hidden] { display: none !important; }`\n",
       "     so we also need the `!important` here to be able to override the\n",
       "     default hidden behavior on the sphinx rendered scikit-learn.org.\n",
       "     See: https://github.com/scikit-learn/scikit-learn/issues/21755 */\n",
       "  display: inline-block !important;\n",
       "  position: relative;\n",
       "}\n",
       "\n",
       "#sk-container-id-2 div.sk-text-repr-fallback {\n",
       "  display: none;\n",
       "}\n",
       "\n",
       "div.sk-parallel-item,\n",
       "div.sk-serial,\n",
       "div.sk-item {\n",
       "  /* draw centered vertical line to link estimators */\n",
       "  background-image: linear-gradient(var(--sklearn-color-text-on-default-background), var(--sklearn-color-text-on-default-background));\n",
       "  background-size: 2px 100%;\n",
       "  background-repeat: no-repeat;\n",
       "  background-position: center center;\n",
       "}\n",
       "\n",
       "/* Parallel-specific style estimator block */\n",
       "\n",
       "#sk-container-id-2 div.sk-parallel-item::after {\n",
       "  content: \"\";\n",
       "  width: 100%;\n",
       "  border-bottom: 2px solid var(--sklearn-color-text-on-default-background);\n",
       "  flex-grow: 1;\n",
       "}\n",
       "\n",
       "#sk-container-id-2 div.sk-parallel {\n",
       "  display: flex;\n",
       "  align-items: stretch;\n",
       "  justify-content: center;\n",
       "  background-color: var(--sklearn-color-background);\n",
       "  position: relative;\n",
       "}\n",
       "\n",
       "#sk-container-id-2 div.sk-parallel-item {\n",
       "  display: flex;\n",
       "  flex-direction: column;\n",
       "}\n",
       "\n",
       "#sk-container-id-2 div.sk-parallel-item:first-child::after {\n",
       "  align-self: flex-end;\n",
       "  width: 50%;\n",
       "}\n",
       "\n",
       "#sk-container-id-2 div.sk-parallel-item:last-child::after {\n",
       "  align-self: flex-start;\n",
       "  width: 50%;\n",
       "}\n",
       "\n",
       "#sk-container-id-2 div.sk-parallel-item:only-child::after {\n",
       "  width: 0;\n",
       "}\n",
       "\n",
       "/* Serial-specific style estimator block */\n",
       "\n",
       "#sk-container-id-2 div.sk-serial {\n",
       "  display: flex;\n",
       "  flex-direction: column;\n",
       "  align-items: center;\n",
       "  background-color: var(--sklearn-color-background);\n",
       "  padding-right: 1em;\n",
       "  padding-left: 1em;\n",
       "}\n",
       "\n",
       "\n",
       "/* Toggleable style: style used for estimator/Pipeline/ColumnTransformer box that is\n",
       "clickable and can be expanded/collapsed.\n",
       "- Pipeline and ColumnTransformer use this feature and define the default style\n",
       "- Estimators will overwrite some part of the style using the `sk-estimator` class\n",
       "*/\n",
       "\n",
       "/* Pipeline and ColumnTransformer style (default) */\n",
       "\n",
       "#sk-container-id-2 div.sk-toggleable {\n",
       "  /* Default theme specific background. It is overwritten whether we have a\n",
       "  specific estimator or a Pipeline/ColumnTransformer */\n",
       "  background-color: var(--sklearn-color-background);\n",
       "}\n",
       "\n",
       "/* Toggleable label */\n",
       "#sk-container-id-2 label.sk-toggleable__label {\n",
       "  cursor: pointer;\n",
       "  display: flex;\n",
       "  width: 100%;\n",
       "  margin-bottom: 0;\n",
       "  padding: 0.5em;\n",
       "  box-sizing: border-box;\n",
       "  text-align: center;\n",
       "  align-items: center;\n",
       "  justify-content: center;\n",
       "  gap: 0.5em;\n",
       "}\n",
       "\n",
       "#sk-container-id-2 label.sk-toggleable__label .caption {\n",
       "  font-size: 0.6rem;\n",
       "  font-weight: lighter;\n",
       "  color: var(--sklearn-color-text-muted);\n",
       "}\n",
       "\n",
       "#sk-container-id-2 label.sk-toggleable__label-arrow:before {\n",
       "  /* Arrow on the left of the label */\n",
       "  content: \"▸\";\n",
       "  float: left;\n",
       "  margin-right: 0.25em;\n",
       "  color: var(--sklearn-color-icon);\n",
       "}\n",
       "\n",
       "#sk-container-id-2 label.sk-toggleable__label-arrow:hover:before {\n",
       "  color: var(--sklearn-color-text);\n",
       "}\n",
       "\n",
       "/* Toggleable content - dropdown */\n",
       "\n",
       "#sk-container-id-2 div.sk-toggleable__content {\n",
       "  display: none;\n",
       "  text-align: left;\n",
       "  /* unfitted */\n",
       "  background-color: var(--sklearn-color-unfitted-level-0);\n",
       "}\n",
       "\n",
       "#sk-container-id-2 div.sk-toggleable__content.fitted {\n",
       "  /* fitted */\n",
       "  background-color: var(--sklearn-color-fitted-level-0);\n",
       "}\n",
       "\n",
       "#sk-container-id-2 div.sk-toggleable__content pre {\n",
       "  margin: 0.2em;\n",
       "  border-radius: 0.25em;\n",
       "  color: var(--sklearn-color-text);\n",
       "  /* unfitted */\n",
       "  background-color: var(--sklearn-color-unfitted-level-0);\n",
       "}\n",
       "\n",
       "#sk-container-id-2 div.sk-toggleable__content.fitted pre {\n",
       "  /* unfitted */\n",
       "  background-color: var(--sklearn-color-fitted-level-0);\n",
       "}\n",
       "\n",
       "#sk-container-id-2 input.sk-toggleable__control:checked~div.sk-toggleable__content {\n",
       "  /* Expand drop-down */\n",
       "  display: block;\n",
       "  width: 100%;\n",
       "  overflow: visible;\n",
       "}\n",
       "\n",
       "#sk-container-id-2 input.sk-toggleable__control:checked~label.sk-toggleable__label-arrow:before {\n",
       "  content: \"▾\";\n",
       "}\n",
       "\n",
       "/* Pipeline/ColumnTransformer-specific style */\n",
       "\n",
       "#sk-container-id-2 div.sk-label input.sk-toggleable__control:checked~label.sk-toggleable__label {\n",
       "  color: var(--sklearn-color-text);\n",
       "  background-color: var(--sklearn-color-unfitted-level-2);\n",
       "}\n",
       "\n",
       "#sk-container-id-2 div.sk-label.fitted input.sk-toggleable__control:checked~label.sk-toggleable__label {\n",
       "  background-color: var(--sklearn-color-fitted-level-2);\n",
       "}\n",
       "\n",
       "/* Estimator-specific style */\n",
       "\n",
       "/* Colorize estimator box */\n",
       "#sk-container-id-2 div.sk-estimator input.sk-toggleable__control:checked~label.sk-toggleable__label {\n",
       "  /* unfitted */\n",
       "  background-color: var(--sklearn-color-unfitted-level-2);\n",
       "}\n",
       "\n",
       "#sk-container-id-2 div.sk-estimator.fitted input.sk-toggleable__control:checked~label.sk-toggleable__label {\n",
       "  /* fitted */\n",
       "  background-color: var(--sklearn-color-fitted-level-2);\n",
       "}\n",
       "\n",
       "#sk-container-id-2 div.sk-label label.sk-toggleable__label,\n",
       "#sk-container-id-2 div.sk-label label {\n",
       "  /* The background is the default theme color */\n",
       "  color: var(--sklearn-color-text-on-default-background);\n",
       "}\n",
       "\n",
       "/* On hover, darken the color of the background */\n",
       "#sk-container-id-2 div.sk-label:hover label.sk-toggleable__label {\n",
       "  color: var(--sklearn-color-text);\n",
       "  background-color: var(--sklearn-color-unfitted-level-2);\n",
       "}\n",
       "\n",
       "/* Label box, darken color on hover, fitted */\n",
       "#sk-container-id-2 div.sk-label.fitted:hover label.sk-toggleable__label.fitted {\n",
       "  color: var(--sklearn-color-text);\n",
       "  background-color: var(--sklearn-color-fitted-level-2);\n",
       "}\n",
       "\n",
       "/* Estimator label */\n",
       "\n",
       "#sk-container-id-2 div.sk-label label {\n",
       "  font-family: monospace;\n",
       "  font-weight: bold;\n",
       "  line-height: 1.2em;\n",
       "}\n",
       "\n",
       "#sk-container-id-2 div.sk-label-container {\n",
       "  text-align: center;\n",
       "}\n",
       "\n",
       "/* Estimator-specific */\n",
       "#sk-container-id-2 div.sk-estimator {\n",
       "  font-family: monospace;\n",
       "  border: 1px dotted var(--sklearn-color-border-box);\n",
       "  border-radius: 0.25em;\n",
       "  box-sizing: border-box;\n",
       "  margin-bottom: 0.5em;\n",
       "  /* unfitted */\n",
       "  background-color: var(--sklearn-color-unfitted-level-0);\n",
       "}\n",
       "\n",
       "#sk-container-id-2 div.sk-estimator.fitted {\n",
       "  /* fitted */\n",
       "  background-color: var(--sklearn-color-fitted-level-0);\n",
       "}\n",
       "\n",
       "/* on hover */\n",
       "#sk-container-id-2 div.sk-estimator:hover {\n",
       "  /* unfitted */\n",
       "  background-color: var(--sklearn-color-unfitted-level-2);\n",
       "}\n",
       "\n",
       "#sk-container-id-2 div.sk-estimator.fitted:hover {\n",
       "  /* fitted */\n",
       "  background-color: var(--sklearn-color-fitted-level-2);\n",
       "}\n",
       "\n",
       "/* Specification for estimator info (e.g. \"i\" and \"?\") */\n",
       "\n",
       "/* Common style for \"i\" and \"?\" */\n",
       "\n",
       ".sk-estimator-doc-link,\n",
       "a:link.sk-estimator-doc-link,\n",
       "a:visited.sk-estimator-doc-link {\n",
       "  float: right;\n",
       "  font-size: smaller;\n",
       "  line-height: 1em;\n",
       "  font-family: monospace;\n",
       "  background-color: var(--sklearn-color-unfitted-level-0);\n",
       "  border-radius: 1em;\n",
       "  height: 1em;\n",
       "  width: 1em;\n",
       "  text-decoration: none !important;\n",
       "  margin-left: 0.5em;\n",
       "  text-align: center;\n",
       "  /* unfitted */\n",
       "  border: var(--sklearn-color-unfitted-level-3) 1pt solid;\n",
       "  color: var(--sklearn-color-unfitted-level-3);\n",
       "}\n",
       "\n",
       ".sk-estimator-doc-link.fitted,\n",
       "a:link.sk-estimator-doc-link.fitted,\n",
       "a:visited.sk-estimator-doc-link.fitted {\n",
       "  /* fitted */\n",
       "  background-color: var(--sklearn-color-fitted-level-0);\n",
       "  border: var(--sklearn-color-fitted-level-3) 1pt solid;\n",
       "  color: var(--sklearn-color-fitted-level-3);\n",
       "}\n",
       "\n",
       "/* On hover */\n",
       "div.sk-estimator:hover .sk-estimator-doc-link:hover,\n",
       ".sk-estimator-doc-link:hover,\n",
       "div.sk-label-container:hover .sk-estimator-doc-link:hover,\n",
       ".sk-estimator-doc-link:hover {\n",
       "  /* unfitted */\n",
       "  background-color: var(--sklearn-color-unfitted-level-3);\n",
       "  border: var(--sklearn-color-fitted-level-0) 1pt solid;\n",
       "  color: var(--sklearn-color-unfitted-level-0);\n",
       "  text-decoration: none;\n",
       "}\n",
       "\n",
       "div.sk-estimator.fitted:hover .sk-estimator-doc-link.fitted:hover,\n",
       ".sk-estimator-doc-link.fitted:hover,\n",
       "div.sk-label-container:hover .sk-estimator-doc-link.fitted:hover,\n",
       ".sk-estimator-doc-link.fitted:hover {\n",
       "  /* fitted */\n",
       "  background-color: var(--sklearn-color-fitted-level-3);\n",
       "  border: var(--sklearn-color-fitted-level-0) 1pt solid;\n",
       "  color: var(--sklearn-color-fitted-level-0);\n",
       "  text-decoration: none;\n",
       "}\n",
       "\n",
       "/* Span, style for the box shown on hovering the info icon */\n",
       ".sk-estimator-doc-link span {\n",
       "  display: none;\n",
       "  z-index: 9999;\n",
       "  position: relative;\n",
       "  font-weight: normal;\n",
       "  right: .2ex;\n",
       "  padding: .5ex;\n",
       "  margin: .5ex;\n",
       "  width: min-content;\n",
       "  min-width: 20ex;\n",
       "  max-width: 50ex;\n",
       "  color: var(--sklearn-color-text);\n",
       "  box-shadow: 2pt 2pt 4pt #999;\n",
       "  /* unfitted */\n",
       "  background: var(--sklearn-color-unfitted-level-0);\n",
       "  border: .5pt solid var(--sklearn-color-unfitted-level-3);\n",
       "}\n",
       "\n",
       ".sk-estimator-doc-link.fitted span {\n",
       "  /* fitted */\n",
       "  background: var(--sklearn-color-fitted-level-0);\n",
       "  border: var(--sklearn-color-fitted-level-3);\n",
       "}\n",
       "\n",
       ".sk-estimator-doc-link:hover span {\n",
       "  display: block;\n",
       "}\n",
       "\n",
       "/* \"?\"-specific style due to the `<a>` HTML tag */\n",
       "\n",
       "#sk-container-id-2 a.estimator_doc_link {\n",
       "  float: right;\n",
       "  font-size: 1rem;\n",
       "  line-height: 1em;\n",
       "  font-family: monospace;\n",
       "  background-color: var(--sklearn-color-unfitted-level-0);\n",
       "  border-radius: 1rem;\n",
       "  height: 1rem;\n",
       "  width: 1rem;\n",
       "  text-decoration: none;\n",
       "  /* unfitted */\n",
       "  color: var(--sklearn-color-unfitted-level-1);\n",
       "  border: var(--sklearn-color-unfitted-level-1) 1pt solid;\n",
       "}\n",
       "\n",
       "#sk-container-id-2 a.estimator_doc_link.fitted {\n",
       "  /* fitted */\n",
       "  background-color: var(--sklearn-color-fitted-level-0);\n",
       "  border: var(--sklearn-color-fitted-level-1) 1pt solid;\n",
       "  color: var(--sklearn-color-fitted-level-1);\n",
       "}\n",
       "\n",
       "/* On hover */\n",
       "#sk-container-id-2 a.estimator_doc_link:hover {\n",
       "  /* unfitted */\n",
       "  background-color: var(--sklearn-color-unfitted-level-3);\n",
       "  color: var(--sklearn-color-background);\n",
       "  text-decoration: none;\n",
       "}\n",
       "\n",
       "#sk-container-id-2 a.estimator_doc_link.fitted:hover {\n",
       "  /* fitted */\n",
       "  background-color: var(--sklearn-color-fitted-level-3);\n",
       "}\n",
       "\n",
       ".estimator-table {\n",
       "    font-family: monospace;\n",
       "}\n",
       "\n",
       ".estimator-table summary {\n",
       "    padding: .5rem;\n",
       "    cursor: pointer;\n",
       "}\n",
       "\n",
       ".estimator-table summary::marker {\n",
       "    font-size: 0.7rem;\n",
       "}\n",
       "\n",
       ".estimator-table details[open] {\n",
       "    padding-left: 0.1rem;\n",
       "    padding-right: 0.1rem;\n",
       "    padding-bottom: 0.3rem;\n",
       "}\n",
       "\n",
       ".estimator-table .parameters-table {\n",
       "    margin-left: auto !important;\n",
       "    margin-right: auto !important;\n",
       "    margin-top: 0;\n",
       "}\n",
       "\n",
       ".estimator-table .parameters-table tr:nth-child(odd) {\n",
       "    background-color: #fff;\n",
       "}\n",
       "\n",
       ".estimator-table .parameters-table tr:nth-child(even) {\n",
       "    background-color: #f6f6f6;\n",
       "}\n",
       "\n",
       ".estimator-table .parameters-table tr:hover {\n",
       "    background-color: #e0e0e0;\n",
       "}\n",
       "\n",
       ".estimator-table table td {\n",
       "    border: 1px solid rgba(106, 105, 104, 0.232);\n",
       "}\n",
       "\n",
       "/*\n",
       "    `table td`is set in notebook with right text-align.\n",
       "    We need to overwrite it.\n",
       "*/\n",
       ".estimator-table table td.param {\n",
       "    text-align: left;\n",
       "    position: relative;\n",
       "    padding: 0;\n",
       "}\n",
       "\n",
       ".user-set td {\n",
       "    color:rgb(255, 94, 0);\n",
       "    text-align: left !important;\n",
       "}\n",
       "\n",
       ".user-set td.value {\n",
       "    color:rgb(255, 94, 0);\n",
       "    background-color: transparent;\n",
       "}\n",
       "\n",
       ".default td {\n",
       "    color: black;\n",
       "    text-align: left !important;\n",
       "}\n",
       "\n",
       ".user-set td i,\n",
       ".default td i {\n",
       "    color: black;\n",
       "}\n",
       "\n",
       "/*\n",
       "    Styles for parameter documentation links\n",
       "    We need styling for visited so jupyter doesn't overwrite it\n",
       "*/\n",
       "a.param-doc-link,\n",
       "a.param-doc-link:link,\n",
       "a.param-doc-link:visited {\n",
       "    text-decoration: underline dashed;\n",
       "    text-underline-offset: .3em;\n",
       "    color: inherit;\n",
       "    display: block;\n",
       "    padding: .5em;\n",
       "}\n",
       "\n",
       "/* \"hack\" to make the entire area of the cell containing the link clickable */\n",
       "a.param-doc-link::before {\n",
       "    position: absolute;\n",
       "    content: \"\";\n",
       "    inset: 0;\n",
       "}\n",
       "\n",
       ".param-doc-description {\n",
       "    display: none;\n",
       "    position: absolute;\n",
       "    z-index: 9999;\n",
       "    left: 0;\n",
       "    padding: .5ex;\n",
       "    margin-left: 1.5em;\n",
       "    color: var(--sklearn-color-text);\n",
       "    box-shadow: .3em .3em .4em #999;\n",
       "    width: max-content;\n",
       "    text-align: left;\n",
       "    max-height: 10em;\n",
       "    overflow-y: auto;\n",
       "\n",
       "    /* unfitted */\n",
       "    background: var(--sklearn-color-unfitted-level-0);\n",
       "    border: thin solid var(--sklearn-color-unfitted-level-3);\n",
       "}\n",
       "\n",
       "/* Fitted state for parameter tooltips */\n",
       ".fitted .param-doc-description {\n",
       "    /* fitted */\n",
       "    background: var(--sklearn-color-fitted-level-0);\n",
       "    border: thin solid var(--sklearn-color-fitted-level-3);\n",
       "}\n",
       "\n",
       ".param-doc-link:hover .param-doc-description {\n",
       "    display: block;\n",
       "}\n",
       "\n",
       ".copy-paste-icon {\n",
       "    background-image: url(data:image/svg+xml;base64,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);\n",
       "    background-repeat: no-repeat;\n",
       "    background-size: 14px 14px;\n",
       "    background-position: 0;\n",
       "    display: inline-block;\n",
       "    width: 14px;\n",
       "    height: 14px;\n",
       "    cursor: pointer;\n",
       "}\n",
       "</style><body><div id=\"sk-container-id-2\" class=\"sk-top-container\"><div class=\"sk-text-repr-fallback\"><pre>XGBRegressor(base_score=np.float64(0.013271975120564444), booster=None,\n",
       "             callbacks=None, colsample_bylevel=None, colsample_bynode=None,\n",
       "             colsample_bytree=0.5, device=None, early_stopping_rounds=20,\n",
       "             enable_categorical=False, eval_metric=None, feature_types=None,\n",
       "             feature_weights=None, gamma=None, grow_policy=None,\n",
       "             importance_type=None, interaction_constraints=None,\n",
       "             learning_rate=0.02, max_bin=None, max_cat_threshold=None,\n",
       "             max_cat_to_onehot=None, max_delta_step=None, max_depth=None,\n",
       "             max_leaves=None, min_child_weight=None, missing=nan,\n",
       "             monotone_constraints=None, multi_strategy=None, n_estimators=5000,\n",
       "             n_jobs=None, num_parallel_tree=None, ...)</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\"><div class=\"sk-estimator fitted sk-toggleable\"><input class=\"sk-toggleable__control sk-hidden--visually\" id=\"sk-estimator-id-2\" type=\"checkbox\" checked><label for=\"sk-estimator-id-2\" class=\"sk-toggleable__label fitted sk-toggleable__label-arrow\"><div><div>XGBRegressor</div></div><div><a class=\"sk-estimator-doc-link fitted\" rel=\"noreferrer\" target=\"_blank\" href=\"https://xgboost.readthedocs.io/en/release_3.2.0/python/python_api.html#xgboost.XGBRegressor\">?<span>Documentation for XGBRegressor</span></a><span class=\"sk-estimator-doc-link fitted\">i<span>Fitted</span></span></div></label><div class=\"sk-toggleable__content fitted\" data-param-prefix=\"\">\n",
       "        <div class=\"estimator-table\">\n",
       "            <details>\n",
       "                <summary>Parameters</summary>\n",
       "                <table class=\"parameters-table\">\n",
       "                  <tbody>\n",
       "                    \n",
       "        <tr class=\"default\">\n",
       "            <td><i class=\"copy-paste-icon\"\n",
       "                 onclick=\"copyToClipboard('objective',\n",
       "                          this.parentElement.nextElementSibling)\"\n",
       "            ></i></td>\n",
       "            <td class=\"param\">\n",
       "        <a class=\"param-doc-link\"\n",
       "            rel=\"noreferrer\" target=\"_blank\" href=\"https://xgboost.readthedocs.io/en/release_3.2.0/python/python_api.html#xgboost.XGBRegressor#:~:text=objective,-typing.Union%5Bstr%2C%20xgboost.sklearn._SklObjWProto%2C%20typing.Callable%5B%5Btyping.Any%2C%20typing.Any%5D%2C%20typing.Tuple%5Bnumpy.ndarray%2C%20numpy.ndarray%5D%5D%2C%20NoneType%5D\">\n",
       "            objective\n",
       "            <span class=\"param-doc-description\">objective: typing.Union[str, xgboost.sklearn._SklObjWProto, typing.Callable[[typing.Any, typing.Any], typing.Tuple[numpy.ndarray, numpy.ndarray]], NoneType]<br><br>Specify the learning task and the corresponding learning objective or a custom<br>objective function to be used.<br><br>For custom objective, see :doc:`/tutorials/custom_metric_obj` and<br>:ref:`custom-obj-metric` for more information, along with the end note for<br>function signatures.</span>\n",
       "        </a>\n",
       "    </td>\n",
       "            <td class=\"value\">&#x27;reg:squarederror&#x27;</td>\n",
       "        </tr>\n",
       "    \n",
       "\n",
       "        <tr class=\"user-set\">\n",
       "            <td><i class=\"copy-paste-icon\"\n",
       "                 onclick=\"copyToClipboard('base_score',\n",
       "                          this.parentElement.nextElementSibling)\"\n",
       "            ></i></td>\n",
       "            <td class=\"param\">\n",
       "        <a class=\"param-doc-link\"\n",
       "            rel=\"noreferrer\" target=\"_blank\" href=\"https://xgboost.readthedocs.io/en/release_3.2.0/python/python_api.html#xgboost.XGBRegressor#:~:text=base_score,-typing.Union%5Bfloat%2C%20typing.List%5Bfloat%5D%2C%20NoneType%5D\">\n",
       "            base_score\n",
       "            <span class=\"param-doc-description\">base_score: typing.Union[float, typing.List[float], NoneType]<br><br>The initial prediction score of all instances, global bias.</span>\n",
       "        </a>\n",
       "    </td>\n",
       "            <td class=\"value\">np.float64(0....1975120564444)</td>\n",
       "        </tr>\n",
       "    \n",
       "\n",
       "        <tr class=\"user-set\">\n",
       "            <td><i class=\"copy-paste-icon\"\n",
       "                 onclick=\"copyToClipboard('booster',\n",
       "                          this.parentElement.nextElementSibling)\"\n",
       "            ></i></td>\n",
       "            <td class=\"param\">booster</td>\n",
       "            <td class=\"value\">None</td>\n",
       "        </tr>\n",
       "    \n",
       "\n",
       "        <tr class=\"user-set\">\n",
       "            <td><i class=\"copy-paste-icon\"\n",
       "                 onclick=\"copyToClipboard('callbacks',\n",
       "                          this.parentElement.nextElementSibling)\"\n",
       "            ></i></td>\n",
       "            <td class=\"param\">\n",
       "        <a class=\"param-doc-link\"\n",
       "            rel=\"noreferrer\" target=\"_blank\" href=\"https://xgboost.readthedocs.io/en/release_3.2.0/python/python_api.html#xgboost.XGBRegressor#:~:text=callbacks,-typing.Optional%5Btyping.List%5Bxgboost.callback.TrainingCallback%5D%5D\">\n",
       "            callbacks\n",
       "            <span class=\"param-doc-description\">callbacks: typing.Optional[typing.List[xgboost.callback.TrainingCallback]]<br><br>List of callback functions that are applied at end of each iteration.<br>It is possible to use predefined callbacks by using<br>:ref:`Callback API <callback_api>`.<br><br>.. note::<br><br>   States in callback are not preserved during training, which means callback<br>   objects can not be reused for multiple training sessions without<br>   reinitialization or deepcopy.<br><br>.. code-block:: python<br><br>    for params in parameters_grid:<br>        # be sure to (re)initialize the callbacks before each run<br>        callbacks = [xgb.callback.LearningRateScheduler(custom_rates)]<br>        reg = xgboost.XGBRegressor(**params, callbacks=callbacks)<br>        reg.fit(X, y)</span>\n",
       "        </a>\n",
       "    </td>\n",
       "            <td class=\"value\">None</td>\n",
       "        </tr>\n",
       "    \n",
       "\n",
       "        <tr class=\"user-set\">\n",
       "            <td><i class=\"copy-paste-icon\"\n",
       "                 onclick=\"copyToClipboard('colsample_bylevel',\n",
       "                          this.parentElement.nextElementSibling)\"\n",
       "            ></i></td>\n",
       "            <td class=\"param\">\n",
       "        <a class=\"param-doc-link\"\n",
       "            rel=\"noreferrer\" target=\"_blank\" href=\"https://xgboost.readthedocs.io/en/release_3.2.0/python/python_api.html#xgboost.XGBRegressor#:~:text=colsample_bylevel,-typing.Optional%5Bfloat%5D\">\n",
       "            colsample_bylevel\n",
       "            <span class=\"param-doc-description\">colsample_bylevel: typing.Optional[float]<br><br>Subsample ratio of columns for each level.</span>\n",
       "        </a>\n",
       "    </td>\n",
       "            <td class=\"value\">None</td>\n",
       "        </tr>\n",
       "    \n",
       "\n",
       "        <tr class=\"user-set\">\n",
       "            <td><i class=\"copy-paste-icon\"\n",
       "                 onclick=\"copyToClipboard('colsample_bynode',\n",
       "                          this.parentElement.nextElementSibling)\"\n",
       "            ></i></td>\n",
       "            <td class=\"param\">\n",
       "        <a class=\"param-doc-link\"\n",
       "            rel=\"noreferrer\" target=\"_blank\" href=\"https://xgboost.readthedocs.io/en/release_3.2.0/python/python_api.html#xgboost.XGBRegressor#:~:text=colsample_bynode,-typing.Optional%5Bfloat%5D\">\n",
       "            colsample_bynode\n",
       "            <span class=\"param-doc-description\">colsample_bynode: typing.Optional[float]<br><br>Subsample ratio of columns for each split.</span>\n",
       "        </a>\n",
       "    </td>\n",
       "            <td class=\"value\">None</td>\n",
       "        </tr>\n",
       "    \n",
       "\n",
       "        <tr class=\"user-set\">\n",
       "            <td><i class=\"copy-paste-icon\"\n",
       "                 onclick=\"copyToClipboard('colsample_bytree',\n",
       "                          this.parentElement.nextElementSibling)\"\n",
       "            ></i></td>\n",
       "            <td class=\"param\">\n",
       "        <a class=\"param-doc-link\"\n",
       "            rel=\"noreferrer\" target=\"_blank\" href=\"https://xgboost.readthedocs.io/en/release_3.2.0/python/python_api.html#xgboost.XGBRegressor#:~:text=colsample_bytree,-typing.Optional%5Bfloat%5D\">\n",
       "            colsample_bytree\n",
       "            <span class=\"param-doc-description\">colsample_bytree: typing.Optional[float]<br><br>Subsample ratio of columns when constructing each tree.</span>\n",
       "        </a>\n",
       "    </td>\n",
       "            <td class=\"value\">0.5</td>\n",
       "        </tr>\n",
       "    \n",
       "\n",
       "        <tr class=\"user-set\">\n",
       "            <td><i class=\"copy-paste-icon\"\n",
       "                 onclick=\"copyToClipboard('device',\n",
       "                          this.parentElement.nextElementSibling)\"\n",
       "            ></i></td>\n",
       "            <td class=\"param\">\n",
       "        <a class=\"param-doc-link\"\n",
       "            rel=\"noreferrer\" target=\"_blank\" href=\"https://xgboost.readthedocs.io/en/release_3.2.0/python/python_api.html#xgboost.XGBRegressor#:~:text=device,-typing.Optional%5Bstr%5D\">\n",
       "            device\n",
       "            <span class=\"param-doc-description\">device: typing.Optional[str]<br><br>.. versionadded:: 2.0.0<br><br>Device ordinal, available options are `cpu`, `cuda`, and `gpu`.</span>\n",
       "        </a>\n",
       "    </td>\n",
       "            <td class=\"value\">None</td>\n",
       "        </tr>\n",
       "    \n",
       "\n",
       "        <tr class=\"user-set\">\n",
       "            <td><i class=\"copy-paste-icon\"\n",
       "                 onclick=\"copyToClipboard('early_stopping_rounds',\n",
       "                          this.parentElement.nextElementSibling)\"\n",
       "            ></i></td>\n",
       "            <td class=\"param\">\n",
       "        <a class=\"param-doc-link\"\n",
       "            rel=\"noreferrer\" target=\"_blank\" href=\"https://xgboost.readthedocs.io/en/release_3.2.0/python/python_api.html#xgboost.XGBRegressor#:~:text=early_stopping_rounds,-typing.Optional%5Bint%5D\">\n",
       "            early_stopping_rounds\n",
       "            <span class=\"param-doc-description\">early_stopping_rounds: typing.Optional[int]<br><br>.. versionadded:: 1.6.0<br><br>- Activates early stopping. Validation metric needs to improve at least once in<br>  every **early_stopping_rounds** round(s) to continue training.  Requires at<br>  least one item in **eval_set** in :py:meth:`fit`.<br><br>- If early stopping occurs, the model will have two additional attributes:<br>  :py:attr:`best_score` and :py:attr:`best_iteration`. These are used by the<br>  :py:meth:`predict` and :py:meth:`apply` methods to determine the optimal<br>  number of trees during inference. If users want to access the full model<br>  (including trees built after early stopping), they can specify the<br>  `iteration_range` in these inference methods. In addition, other utilities<br>  like model plotting can also use the entire model.<br><br>- If you prefer to discard the trees after `best_iteration`, consider using the<br>  callback function :py:class:`xgboost.callback.EarlyStopping`.<br><br>- If there's more than one item in **eval_set**, the last entry will be used for<br>  early stopping.  If there's more than one metric in **eval_metric**, the last<br>  metric will be used for early stopping.</span>\n",
       "        </a>\n",
       "    </td>\n",
       "            <td class=\"value\">20</td>\n",
       "        </tr>\n",
       "    \n",
       "\n",
       "        <tr class=\"user-set\">\n",
       "            <td><i class=\"copy-paste-icon\"\n",
       "                 onclick=\"copyToClipboard('enable_categorical',\n",
       "                          this.parentElement.nextElementSibling)\"\n",
       "            ></i></td>\n",
       "            <td class=\"param\">\n",
       "        <a class=\"param-doc-link\"\n",
       "            rel=\"noreferrer\" target=\"_blank\" href=\"https://xgboost.readthedocs.io/en/release_3.2.0/python/python_api.html#xgboost.XGBRegressor#:~:text=enable_categorical,-bool\">\n",
       "            enable_categorical\n",
       "            <span class=\"param-doc-description\">enable_categorical: bool<br><br>See the same parameter of :py:class:`DMatrix` for details.</span>\n",
       "        </a>\n",
       "    </td>\n",
       "            <td class=\"value\">False</td>\n",
       "        </tr>\n",
       "    \n",
       "\n",
       "        <tr class=\"user-set\">\n",
       "            <td><i class=\"copy-paste-icon\"\n",
       "                 onclick=\"copyToClipboard('eval_metric',\n",
       "                          this.parentElement.nextElementSibling)\"\n",
       "            ></i></td>\n",
       "            <td class=\"param\">\n",
       "        <a class=\"param-doc-link\"\n",
       "            rel=\"noreferrer\" target=\"_blank\" href=\"https://xgboost.readthedocs.io/en/release_3.2.0/python/python_api.html#xgboost.XGBRegressor#:~:text=eval_metric,-typing.Union%5Bstr%2C%20typing.List%5Btyping.Union%5Bstr%2C%20typing.Callable%5D%5D%2C%20typing.Callable%2C%20NoneType%5D\">\n",
       "            eval_metric\n",
       "            <span class=\"param-doc-description\">eval_metric: typing.Union[str, typing.List[typing.Union[str, typing.Callable]], typing.Callable, NoneType]<br><br>.. versionadded:: 1.6.0<br><br>Metric used for monitoring the training result and early stopping.  It can be a<br>string or list of strings as names of predefined metric in XGBoost (See<br>:doc:`/parameter`), one of the metrics in :py:mod:`sklearn.metrics`, or any<br>other user defined metric that looks like `sklearn.metrics`.<br><br>If custom objective is also provided, then custom metric should implement the<br>corresponding reverse link function.<br><br>Unlike the `scoring` parameter commonly used in scikit-learn, when a callable<br>object is provided, it's assumed to be a cost function and by default XGBoost<br>will minimize the result during early stopping.<br><br>For advanced usage on Early stopping like directly choosing to maximize instead<br>of minimize, see :py:obj:`xgboost.callback.EarlyStopping`.<br><br>See :doc:`/tutorials/custom_metric_obj` and :ref:`custom-obj-metric` for more<br>information.<br><br>.. code-block:: python<br><br>    from sklearn.datasets import load_diabetes<br>    from sklearn.metrics import mean_absolute_error<br>    X, y = load_diabetes(return_X_y=True)<br>    reg = xgb.XGBRegressor(<br>        tree_method=\"hist\",<br>        eval_metric=mean_absolute_error,<br>    )<br>    reg.fit(X, y, eval_set=[(X, y)])</span>\n",
       "        </a>\n",
       "    </td>\n",
       "            <td class=\"value\">None</td>\n",
       "        </tr>\n",
       "    \n",
       "\n",
       "        <tr class=\"user-set\">\n",
       "            <td><i class=\"copy-paste-icon\"\n",
       "                 onclick=\"copyToClipboard('feature_types',\n",
       "                          this.parentElement.nextElementSibling)\"\n",
       "            ></i></td>\n",
       "            <td class=\"param\">\n",
       "        <a class=\"param-doc-link\"\n",
       "            rel=\"noreferrer\" target=\"_blank\" href=\"https://xgboost.readthedocs.io/en/release_3.2.0/python/python_api.html#xgboost.XGBRegressor#:~:text=feature_types,-typing.Optional%5Btyping.Sequence%5Bstr%5D%5D\">\n",
       "            feature_types\n",
       "            <span class=\"param-doc-description\">feature_types: typing.Optional[typing.Sequence[str]]<br><br>.. versionadded:: 1.7.0<br><br>Used for specifying feature types without constructing a dataframe. See<br>the :py:class:`DMatrix` for details.</span>\n",
       "        </a>\n",
       "    </td>\n",
       "            <td class=\"value\">None</td>\n",
       "        </tr>\n",
       "    \n",
       "\n",
       "        <tr class=\"user-set\">\n",
       "            <td><i class=\"copy-paste-icon\"\n",
       "                 onclick=\"copyToClipboard('feature_weights',\n",
       "                          this.parentElement.nextElementSibling)\"\n",
       "            ></i></td>\n",
       "            <td class=\"param\">\n",
       "        <a class=\"param-doc-link\"\n",
       "            rel=\"noreferrer\" target=\"_blank\" href=\"https://xgboost.readthedocs.io/en/release_3.2.0/python/python_api.html#xgboost.XGBRegressor#:~:text=feature_weights,-Optional%5BArrayLike%5D\">\n",
       "            feature_weights\n",
       "            <span class=\"param-doc-description\">feature_weights: Optional[ArrayLike]<br><br>Weight for each feature, defines the probability of each feature being selected<br>when colsample is being used.  All values must be greater than 0, otherwise a<br>`ValueError` is thrown.</span>\n",
       "        </a>\n",
       "    </td>\n",
       "            <td class=\"value\">None</td>\n",
       "        </tr>\n",
       "    \n",
       "\n",
       "        <tr class=\"user-set\">\n",
       "            <td><i class=\"copy-paste-icon\"\n",
       "                 onclick=\"copyToClipboard('gamma',\n",
       "                          this.parentElement.nextElementSibling)\"\n",
       "            ></i></td>\n",
       "            <td class=\"param\">\n",
       "        <a class=\"param-doc-link\"\n",
       "            rel=\"noreferrer\" target=\"_blank\" href=\"https://xgboost.readthedocs.io/en/release_3.2.0/python/python_api.html#xgboost.XGBRegressor#:~:text=gamma,-typing.Optional%5Bfloat%5D\">\n",
       "            gamma\n",
       "            <span class=\"param-doc-description\">gamma: typing.Optional[float]<br><br>(min_split_loss) Minimum loss reduction required to make a further partition on<br>a leaf node of the tree.</span>\n",
       "        </a>\n",
       "    </td>\n",
       "            <td class=\"value\">None</td>\n",
       "        </tr>\n",
       "    \n",
       "\n",
       "        <tr class=\"user-set\">\n",
       "            <td><i class=\"copy-paste-icon\"\n",
       "                 onclick=\"copyToClipboard('grow_policy',\n",
       "                          this.parentElement.nextElementSibling)\"\n",
       "            ></i></td>\n",
       "            <td class=\"param\">\n",
       "        <a class=\"param-doc-link\"\n",
       "            rel=\"noreferrer\" target=\"_blank\" href=\"https://xgboost.readthedocs.io/en/release_3.2.0/python/python_api.html#xgboost.XGBRegressor#:~:text=grow_policy,-typing.Optional%5Bstr%5D\">\n",
       "            grow_policy\n",
       "            <span class=\"param-doc-description\">grow_policy: typing.Optional[str]<br><br>Tree growing policy.<br><br>- depthwise: Favors splitting at nodes closest to the node,<br>- lossguide: Favors splitting at nodes with highest loss change.</span>\n",
       "        </a>\n",
       "    </td>\n",
       "            <td class=\"value\">None</td>\n",
       "        </tr>\n",
       "    \n",
       "\n",
       "        <tr class=\"user-set\">\n",
       "            <td><i class=\"copy-paste-icon\"\n",
       "                 onclick=\"copyToClipboard('importance_type',\n",
       "                          this.parentElement.nextElementSibling)\"\n",
       "            ></i></td>\n",
       "            <td class=\"param\">importance_type</td>\n",
       "            <td class=\"value\">None</td>\n",
       "        </tr>\n",
       "    \n",
       "\n",
       "        <tr class=\"user-set\">\n",
       "            <td><i class=\"copy-paste-icon\"\n",
       "                 onclick=\"copyToClipboard('interaction_constraints',\n",
       "                          this.parentElement.nextElementSibling)\"\n",
       "            ></i></td>\n",
       "            <td class=\"param\">\n",
       "        <a class=\"param-doc-link\"\n",
       "            rel=\"noreferrer\" target=\"_blank\" href=\"https://xgboost.readthedocs.io/en/release_3.2.0/python/python_api.html#xgboost.XGBRegressor#:~:text=interaction_constraints,-typing.Union%5Bstr%2C%20typing.List%5Btyping.Tuple%5Bstr%5D%5D%2C%20NoneType%5D\">\n",
       "            interaction_constraints\n",
       "            <span class=\"param-doc-description\">interaction_constraints: typing.Union[str, typing.List[typing.Tuple[str]], NoneType]<br><br>Constraints for interaction representing permitted interactions.  The<br>constraints must be specified in the form of a nested list, e.g. ``[[0, 1], [2,<br>3, 4]]``, where each inner list is a group of indices of features that are<br>allowed to interact with each other.  See :doc:`tutorial<br></tutorials/feature_interaction_constraint>` for more information</span>\n",
       "        </a>\n",
       "    </td>\n",
       "            <td class=\"value\">None</td>\n",
       "        </tr>\n",
       "    \n",
       "\n",
       "        <tr class=\"user-set\">\n",
       "            <td><i class=\"copy-paste-icon\"\n",
       "                 onclick=\"copyToClipboard('learning_rate',\n",
       "                          this.parentElement.nextElementSibling)\"\n",
       "            ></i></td>\n",
       "            <td class=\"param\">\n",
       "        <a class=\"param-doc-link\"\n",
       "            rel=\"noreferrer\" target=\"_blank\" href=\"https://xgboost.readthedocs.io/en/release_3.2.0/python/python_api.html#xgboost.XGBRegressor#:~:text=learning_rate,-typing.Optional%5Bfloat%5D\">\n",
       "            learning_rate\n",
       "            <span class=\"param-doc-description\">learning_rate: typing.Optional[float]<br><br>Boosting learning rate (xgb's \"eta\")</span>\n",
       "        </a>\n",
       "    </td>\n",
       "            <td class=\"value\">0.02</td>\n",
       "        </tr>\n",
       "    \n",
       "\n",
       "        <tr class=\"user-set\">\n",
       "            <td><i class=\"copy-paste-icon\"\n",
       "                 onclick=\"copyToClipboard('max_bin',\n",
       "                          this.parentElement.nextElementSibling)\"\n",
       "            ></i></td>\n",
       "            <td class=\"param\">\n",
       "        <a class=\"param-doc-link\"\n",
       "            rel=\"noreferrer\" target=\"_blank\" href=\"https://xgboost.readthedocs.io/en/release_3.2.0/python/python_api.html#xgboost.XGBRegressor#:~:text=max_bin,-typing.Optional%5Bint%5D\">\n",
       "            max_bin\n",
       "            <span class=\"param-doc-description\">max_bin: typing.Optional[int]<br><br>If using histogram-based algorithm, maximum number of bins per feature</span>\n",
       "        </a>\n",
       "    </td>\n",
       "            <td class=\"value\">None</td>\n",
       "        </tr>\n",
       "    \n",
       "\n",
       "        <tr class=\"user-set\">\n",
       "            <td><i class=\"copy-paste-icon\"\n",
       "                 onclick=\"copyToClipboard('max_cat_threshold',\n",
       "                          this.parentElement.nextElementSibling)\"\n",
       "            ></i></td>\n",
       "            <td class=\"param\">\n",
       "        <a class=\"param-doc-link\"\n",
       "            rel=\"noreferrer\" target=\"_blank\" href=\"https://xgboost.readthedocs.io/en/release_3.2.0/python/python_api.html#xgboost.XGBRegressor#:~:text=max_cat_threshold,-typing.Optional%5Bint%5D\">\n",
       "            max_cat_threshold\n",
       "            <span class=\"param-doc-description\">max_cat_threshold: typing.Optional[int]<br><br>.. versionadded:: 1.7.0<br><br>.. note:: This parameter is experimental<br><br>Maximum number of categories considered for each split. Used only by<br>partition-based splits for preventing over-fitting. Also, `enable_categorical`<br>needs to be set to have categorical feature support. See :doc:`Categorical Data<br></tutorials/categorical>` and :ref:`cat-param` for details.</span>\n",
       "        </a>\n",
       "    </td>\n",
       "            <td class=\"value\">None</td>\n",
       "        </tr>\n",
       "    \n",
       "\n",
       "        <tr class=\"user-set\">\n",
       "            <td><i class=\"copy-paste-icon\"\n",
       "                 onclick=\"copyToClipboard('max_cat_to_onehot',\n",
       "                          this.parentElement.nextElementSibling)\"\n",
       "            ></i></td>\n",
       "            <td class=\"param\">\n",
       "        <a class=\"param-doc-link\"\n",
       "            rel=\"noreferrer\" target=\"_blank\" href=\"https://xgboost.readthedocs.io/en/release_3.2.0/python/python_api.html#xgboost.XGBRegressor#:~:text=max_cat_to_onehot,-Optional%5Bint%5D\">\n",
       "            max_cat_to_onehot\n",
       "            <span class=\"param-doc-description\">max_cat_to_onehot: Optional[int]<br><br>.. versionadded:: 1.6.0<br><br>.. note:: This parameter is experimental<br><br>A threshold for deciding whether XGBoost should use one-hot encoding based split<br>for categorical data.  When number of categories is lesser than the threshold<br>then one-hot encoding is chosen, otherwise the categories will be partitioned<br>into children nodes. Also, `enable_categorical` needs to be set to have<br>categorical feature support. See :doc:`Categorical Data<br></tutorials/categorical>` and :ref:`cat-param` for details.</span>\n",
       "        </a>\n",
       "    </td>\n",
       "            <td class=\"value\">None</td>\n",
       "        </tr>\n",
       "    \n",
       "\n",
       "        <tr class=\"user-set\">\n",
       "            <td><i class=\"copy-paste-icon\"\n",
       "                 onclick=\"copyToClipboard('max_delta_step',\n",
       "                          this.parentElement.nextElementSibling)\"\n",
       "            ></i></td>\n",
       "            <td class=\"param\">\n",
       "        <a class=\"param-doc-link\"\n",
       "            rel=\"noreferrer\" target=\"_blank\" href=\"https://xgboost.readthedocs.io/en/release_3.2.0/python/python_api.html#xgboost.XGBRegressor#:~:text=max_delta_step,-typing.Optional%5Bfloat%5D\">\n",
       "            max_delta_step\n",
       "            <span class=\"param-doc-description\">max_delta_step: typing.Optional[float]<br><br>Maximum delta step we allow each tree's weight estimation to be.</span>\n",
       "        </a>\n",
       "    </td>\n",
       "            <td class=\"value\">None</td>\n",
       "        </tr>\n",
       "    \n",
       "\n",
       "        <tr class=\"user-set\">\n",
       "            <td><i class=\"copy-paste-icon\"\n",
       "                 onclick=\"copyToClipboard('max_depth',\n",
       "                          this.parentElement.nextElementSibling)\"\n",
       "            ></i></td>\n",
       "            <td class=\"param\">\n",
       "        <a class=\"param-doc-link\"\n",
       "            rel=\"noreferrer\" target=\"_blank\" href=\"https://xgboost.readthedocs.io/en/release_3.2.0/python/python_api.html#xgboost.XGBRegressor#:~:text=max_depth,-%20typing.Optional%5Bint%5D\">\n",
       "            max_depth\n",
       "            <span class=\"param-doc-description\">max_depth:  typing.Optional[int]<br><br>Maximum tree depth for base learners.</span>\n",
       "        </a>\n",
       "    </td>\n",
       "            <td class=\"value\">None</td>\n",
       "        </tr>\n",
       "    \n",
       "\n",
       "        <tr class=\"user-set\">\n",
       "            <td><i class=\"copy-paste-icon\"\n",
       "                 onclick=\"copyToClipboard('max_leaves',\n",
       "                          this.parentElement.nextElementSibling)\"\n",
       "            ></i></td>\n",
       "            <td class=\"param\">\n",
       "        <a class=\"param-doc-link\"\n",
       "            rel=\"noreferrer\" target=\"_blank\" href=\"https://xgboost.readthedocs.io/en/release_3.2.0/python/python_api.html#xgboost.XGBRegressor#:~:text=max_leaves,-typing.Optional%5Bint%5D\">\n",
       "            max_leaves\n",
       "            <span class=\"param-doc-description\">max_leaves: typing.Optional[int]<br><br>Maximum number of leaves; 0 indicates no limit.</span>\n",
       "        </a>\n",
       "    </td>\n",
       "            <td class=\"value\">None</td>\n",
       "        </tr>\n",
       "    \n",
       "\n",
       "        <tr class=\"user-set\">\n",
       "            <td><i class=\"copy-paste-icon\"\n",
       "                 onclick=\"copyToClipboard('min_child_weight',\n",
       "                          this.parentElement.nextElementSibling)\"\n",
       "            ></i></td>\n",
       "            <td class=\"param\">\n",
       "        <a class=\"param-doc-link\"\n",
       "            rel=\"noreferrer\" target=\"_blank\" href=\"https://xgboost.readthedocs.io/en/release_3.2.0/python/python_api.html#xgboost.XGBRegressor#:~:text=min_child_weight,-typing.Optional%5Bfloat%5D\">\n",
       "            min_child_weight\n",
       "            <span class=\"param-doc-description\">min_child_weight: typing.Optional[float]<br><br>Minimum sum of instance weight(hessian) needed in a child.</span>\n",
       "        </a>\n",
       "    </td>\n",
       "            <td class=\"value\">None</td>\n",
       "        </tr>\n",
       "    \n",
       "\n",
       "        <tr class=\"user-set\">\n",
       "            <td><i class=\"copy-paste-icon\"\n",
       "                 onclick=\"copyToClipboard('missing',\n",
       "                          this.parentElement.nextElementSibling)\"\n",
       "            ></i></td>\n",
       "            <td class=\"param\">\n",
       "        <a class=\"param-doc-link\"\n",
       "            rel=\"noreferrer\" target=\"_blank\" href=\"https://xgboost.readthedocs.io/en/release_3.2.0/python/python_api.html#xgboost.XGBRegressor#:~:text=missing,-float\">\n",
       "            missing\n",
       "            <span class=\"param-doc-description\">missing: float<br><br>Value in the data which needs to be present as a missing value. Default to<br>:py:data:`numpy.nan`.</span>\n",
       "        </a>\n",
       "    </td>\n",
       "            <td class=\"value\">nan</td>\n",
       "        </tr>\n",
       "    \n",
       "\n",
       "        <tr class=\"user-set\">\n",
       "            <td><i class=\"copy-paste-icon\"\n",
       "                 onclick=\"copyToClipboard('monotone_constraints',\n",
       "                          this.parentElement.nextElementSibling)\"\n",
       "            ></i></td>\n",
       "            <td class=\"param\">\n",
       "        <a class=\"param-doc-link\"\n",
       "            rel=\"noreferrer\" target=\"_blank\" href=\"https://xgboost.readthedocs.io/en/release_3.2.0/python/python_api.html#xgboost.XGBRegressor#:~:text=monotone_constraints,-typing.Union%5Btyping.Dict%5Bstr%2C%20int%5D%2C%20str%2C%20NoneType%5D\">\n",
       "            monotone_constraints\n",
       "            <span class=\"param-doc-description\">monotone_constraints: typing.Union[typing.Dict[str, int], str, NoneType]<br><br>Constraint of variable monotonicity.  See :doc:`tutorial </tutorials/monotonic>`<br>for more information.</span>\n",
       "        </a>\n",
       "    </td>\n",
       "            <td class=\"value\">None</td>\n",
       "        </tr>\n",
       "    \n",
       "\n",
       "        <tr class=\"user-set\">\n",
       "            <td><i class=\"copy-paste-icon\"\n",
       "                 onclick=\"copyToClipboard('multi_strategy',\n",
       "                          this.parentElement.nextElementSibling)\"\n",
       "            ></i></td>\n",
       "            <td class=\"param\">\n",
       "        <a class=\"param-doc-link\"\n",
       "            rel=\"noreferrer\" target=\"_blank\" href=\"https://xgboost.readthedocs.io/en/release_3.2.0/python/python_api.html#xgboost.XGBRegressor#:~:text=multi_strategy,-typing.Optional%5Bstr%5D\">\n",
       "            multi_strategy\n",
       "            <span class=\"param-doc-description\">multi_strategy: typing.Optional[str]<br><br>.. versionadded:: 2.0.0<br><br>.. note:: This parameter is working-in-progress.<br><br>The strategy used for training multi-target models, including multi-target<br>regression and multi-class classification. See :doc:`/tutorials/multioutput` for<br>more information.<br><br>- ``one_output_per_tree``: One model for each target.<br>- ``multi_output_tree``:  Use multi-target trees.</span>\n",
       "        </a>\n",
       "    </td>\n",
       "            <td class=\"value\">None</td>\n",
       "        </tr>\n",
       "    \n",
       "\n",
       "        <tr class=\"user-set\">\n",
       "            <td><i class=\"copy-paste-icon\"\n",
       "                 onclick=\"copyToClipboard('n_estimators',\n",
       "                          this.parentElement.nextElementSibling)\"\n",
       "            ></i></td>\n",
       "            <td class=\"param\">\n",
       "        <a class=\"param-doc-link\"\n",
       "            rel=\"noreferrer\" target=\"_blank\" href=\"https://xgboost.readthedocs.io/en/release_3.2.0/python/python_api.html#xgboost.XGBRegressor#:~:text=n_estimators,-typing.Optional%5Bint%5D\">\n",
       "            n_estimators\n",
       "            <span class=\"param-doc-description\">n_estimators: typing.Optional[int]<br><br>Number of gradient boosted trees.  Equivalent to number of boosting<br>rounds.</span>\n",
       "        </a>\n",
       "    </td>\n",
       "            <td class=\"value\">5000</td>\n",
       "        </tr>\n",
       "    \n",
       "\n",
       "        <tr class=\"user-set\">\n",
       "            <td><i class=\"copy-paste-icon\"\n",
       "                 onclick=\"copyToClipboard('n_jobs',\n",
       "                          this.parentElement.nextElementSibling)\"\n",
       "            ></i></td>\n",
       "            <td class=\"param\">\n",
       "        <a class=\"param-doc-link\"\n",
       "            rel=\"noreferrer\" target=\"_blank\" href=\"https://xgboost.readthedocs.io/en/release_3.2.0/python/python_api.html#xgboost.XGBRegressor#:~:text=n_jobs,-typing.Optional%5Bint%5D\">\n",
       "            n_jobs\n",
       "            <span class=\"param-doc-description\">n_jobs: typing.Optional[int]<br><br>Number of parallel threads used to run xgboost.  When used with other<br>Scikit-Learn algorithms like grid search, you may choose which algorithm to<br>parallelize and balance the threads.  Creating thread contention will<br>significantly slow down both algorithms.</span>\n",
       "        </a>\n",
       "    </td>\n",
       "            <td class=\"value\">None</td>\n",
       "        </tr>\n",
       "    \n",
       "\n",
       "        <tr class=\"user-set\">\n",
       "            <td><i class=\"copy-paste-icon\"\n",
       "                 onclick=\"copyToClipboard('num_parallel_tree',\n",
       "                          this.parentElement.nextElementSibling)\"\n",
       "            ></i></td>\n",
       "            <td class=\"param\">num_parallel_tree</td>\n",
       "            <td class=\"value\">None</td>\n",
       "        </tr>\n",
       "    \n",
       "\n",
       "        <tr class=\"user-set\">\n",
       "            <td><i class=\"copy-paste-icon\"\n",
       "                 onclick=\"copyToClipboard('random_state',\n",
       "                          this.parentElement.nextElementSibling)\"\n",
       "            ></i></td>\n",
       "            <td class=\"param\">\n",
       "        <a class=\"param-doc-link\"\n",
       "            rel=\"noreferrer\" target=\"_blank\" href=\"https://xgboost.readthedocs.io/en/release_3.2.0/python/python_api.html#xgboost.XGBRegressor#:~:text=random_state,-typing.Union%5Bnumpy.random.mtrand.RandomState%2C%20numpy.random._generator.Generator%2C%20int%2C%20NoneType%5D\">\n",
       "            random_state\n",
       "            <span class=\"param-doc-description\">random_state: typing.Union[numpy.random.mtrand.RandomState, numpy.random._generator.Generator, int, NoneType]<br><br>Random number seed.<br><br>.. note::<br><br>   Using gblinear booster with shotgun updater is nondeterministic as<br>   it uses Hogwild algorithm.</span>\n",
       "        </a>\n",
       "    </td>\n",
       "            <td class=\"value\">None</td>\n",
       "        </tr>\n",
       "    \n",
       "\n",
       "        <tr class=\"user-set\">\n",
       "            <td><i class=\"copy-paste-icon\"\n",
       "                 onclick=\"copyToClipboard('reg_alpha',\n",
       "                          this.parentElement.nextElementSibling)\"\n",
       "            ></i></td>\n",
       "            <td class=\"param\">\n",
       "        <a class=\"param-doc-link\"\n",
       "            rel=\"noreferrer\" target=\"_blank\" href=\"https://xgboost.readthedocs.io/en/release_3.2.0/python/python_api.html#xgboost.XGBRegressor#:~:text=reg_alpha,-typing.Optional%5Bfloat%5D\">\n",
       "            reg_alpha\n",
       "            <span class=\"param-doc-description\">reg_alpha: typing.Optional[float]<br><br>L1 regularization term on weights (xgb's alpha).</span>\n",
       "        </a>\n",
       "    </td>\n",
       "            <td class=\"value\">None</td>\n",
       "        </tr>\n",
       "    \n",
       "\n",
       "        <tr class=\"user-set\">\n",
       "            <td><i class=\"copy-paste-icon\"\n",
       "                 onclick=\"copyToClipboard('reg_lambda',\n",
       "                          this.parentElement.nextElementSibling)\"\n",
       "            ></i></td>\n",
       "            <td class=\"param\">\n",
       "        <a class=\"param-doc-link\"\n",
       "            rel=\"noreferrer\" target=\"_blank\" href=\"https://xgboost.readthedocs.io/en/release_3.2.0/python/python_api.html#xgboost.XGBRegressor#:~:text=reg_lambda,-typing.Optional%5Bfloat%5D\">\n",
       "            reg_lambda\n",
       "            <span class=\"param-doc-description\">reg_lambda: typing.Optional[float]<br><br>L2 regularization term on weights (xgb's lambda).</span>\n",
       "        </a>\n",
       "    </td>\n",
       "            <td class=\"value\">None</td>\n",
       "        </tr>\n",
       "    \n",
       "\n",
       "        <tr class=\"user-set\">\n",
       "            <td><i class=\"copy-paste-icon\"\n",
       "                 onclick=\"copyToClipboard('sampling_method',\n",
       "                          this.parentElement.nextElementSibling)\"\n",
       "            ></i></td>\n",
       "            <td class=\"param\">\n",
       "        <a class=\"param-doc-link\"\n",
       "            rel=\"noreferrer\" target=\"_blank\" href=\"https://xgboost.readthedocs.io/en/release_3.2.0/python/python_api.html#xgboost.XGBRegressor#:~:text=sampling_method,-typing.Optional%5Bstr%5D\">\n",
       "            sampling_method\n",
       "            <span class=\"param-doc-description\">sampling_method: typing.Optional[str]<br><br>Sampling method. Used only by the GPU version of ``hist`` tree method.<br><br>- ``uniform``: Select random training instances uniformly.<br>- ``gradient_based``: Select random training instances with higher probability<br>    when the gradient and hessian are larger. (cf. CatBoost)</span>\n",
       "        </a>\n",
       "    </td>\n",
       "            <td class=\"value\">None</td>\n",
       "        </tr>\n",
       "    \n",
       "\n",
       "        <tr class=\"user-set\">\n",
       "            <td><i class=\"copy-paste-icon\"\n",
       "                 onclick=\"copyToClipboard('scale_pos_weight',\n",
       "                          this.parentElement.nextElementSibling)\"\n",
       "            ></i></td>\n",
       "            <td class=\"param\">\n",
       "        <a class=\"param-doc-link\"\n",
       "            rel=\"noreferrer\" target=\"_blank\" href=\"https://xgboost.readthedocs.io/en/release_3.2.0/python/python_api.html#xgboost.XGBRegressor#:~:text=scale_pos_weight,-typing.Optional%5Bfloat%5D\">\n",
       "            scale_pos_weight\n",
       "            <span class=\"param-doc-description\">scale_pos_weight: typing.Optional[float]<br><br>Balancing of positive and negative weights.</span>\n",
       "        </a>\n",
       "    </td>\n",
       "            <td class=\"value\">None</td>\n",
       "        </tr>\n",
       "    \n",
       "\n",
       "        <tr class=\"user-set\">\n",
       "            <td><i class=\"copy-paste-icon\"\n",
       "                 onclick=\"copyToClipboard('subsample',\n",
       "                          this.parentElement.nextElementSibling)\"\n",
       "            ></i></td>\n",
       "            <td class=\"param\">\n",
       "        <a class=\"param-doc-link\"\n",
       "            rel=\"noreferrer\" target=\"_blank\" href=\"https://xgboost.readthedocs.io/en/release_3.2.0/python/python_api.html#xgboost.XGBRegressor#:~:text=subsample,-typing.Optional%5Bfloat%5D\">\n",
       "            subsample\n",
       "            <span class=\"param-doc-description\">subsample: typing.Optional[float]<br><br>Subsample ratio of the training instance.</span>\n",
       "        </a>\n",
       "    </td>\n",
       "            <td class=\"value\">0.2</td>\n",
       "        </tr>\n",
       "    \n",
       "\n",
       "        <tr class=\"user-set\">\n",
       "            <td><i class=\"copy-paste-icon\"\n",
       "                 onclick=\"copyToClipboard('tree_method',\n",
       "                          this.parentElement.nextElementSibling)\"\n",
       "            ></i></td>\n",
       "            <td class=\"param\">\n",
       "        <a class=\"param-doc-link\"\n",
       "            rel=\"noreferrer\" target=\"_blank\" href=\"https://xgboost.readthedocs.io/en/release_3.2.0/python/python_api.html#xgboost.XGBRegressor#:~:text=tree_method,-typing.Optional%5Bstr%5D\">\n",
       "            tree_method\n",
       "            <span class=\"param-doc-description\">tree_method: typing.Optional[str]<br><br>Specify which tree method to use.  Default to auto.  If this parameter is set to<br>default, XGBoost will choose the most conservative option available.  It's<br>recommended to study this option from the parameters document :doc:`tree method<br></treemethod>`</span>\n",
       "        </a>\n",
       "    </td>\n",
       "            <td class=\"value\">None</td>\n",
       "        </tr>\n",
       "    \n",
       "\n",
       "        <tr class=\"user-set\">\n",
       "            <td><i class=\"copy-paste-icon\"\n",
       "                 onclick=\"copyToClipboard('validate_parameters',\n",
       "                          this.parentElement.nextElementSibling)\"\n",
       "            ></i></td>\n",
       "            <td class=\"param\">\n",
       "        <a class=\"param-doc-link\"\n",
       "            rel=\"noreferrer\" target=\"_blank\" href=\"https://xgboost.readthedocs.io/en/release_3.2.0/python/python_api.html#xgboost.XGBRegressor#:~:text=validate_parameters,-typing.Optional%5Bbool%5D\">\n",
       "            validate_parameters\n",
       "            <span class=\"param-doc-description\">validate_parameters: typing.Optional[bool]<br><br>Give warnings for unknown parameter.</span>\n",
       "        </a>\n",
       "    </td>\n",
       "            <td class=\"value\">None</td>\n",
       "        </tr>\n",
       "    \n",
       "\n",
       "        <tr class=\"user-set\">\n",
       "            <td><i class=\"copy-paste-icon\"\n",
       "                 onclick=\"copyToClipboard('verbosity',\n",
       "                          this.parentElement.nextElementSibling)\"\n",
       "            ></i></td>\n",
       "            <td class=\"param\">\n",
       "        <a class=\"param-doc-link\"\n",
       "            rel=\"noreferrer\" target=\"_blank\" href=\"https://xgboost.readthedocs.io/en/release_3.2.0/python/python_api.html#xgboost.XGBRegressor#:~:text=verbosity,-typing.Optional%5Bint%5D\">\n",
       "            verbosity\n",
       "            <span class=\"param-doc-description\">verbosity: typing.Optional[int]<br><br>The degree of verbosity. Valid values are 0 (silent) - 3 (debug).</span>\n",
       "        </a>\n",
       "    </td>\n",
       "            <td class=\"value\">None</td>\n",
       "        </tr>\n",
       "    \n",
       "                  </tbody>\n",
       "                </table>\n",
       "            </details>\n",
       "        </div>\n",
       "    </div></div></div></div></div><script>function copyToClipboard(text, element) {\n",
       "    // Get the parameter prefix from the closest toggleable content\n",
       "    const toggleableContent = element.closest('.sk-toggleable__content');\n",
       "    const paramPrefix = toggleableContent ? toggleableContent.dataset.paramPrefix : '';\n",
       "    const fullParamName = paramPrefix ? `${paramPrefix}${text}` : text;\n",
       "\n",
       "    const originalStyle = element.style;\n",
       "    const computedStyle = window.getComputedStyle(element);\n",
       "    const originalWidth = computedStyle.width;\n",
       "    const originalHTML = element.innerHTML.replace('Copied!', '');\n",
       "\n",
       "    navigator.clipboard.writeText(fullParamName)\n",
       "        .then(() => {\n",
       "            element.style.width = originalWidth;\n",
       "            element.style.color = 'green';\n",
       "            element.innerHTML = \"Copied!\";\n",
       "\n",
       "            setTimeout(() => {\n",
       "                element.innerHTML = originalHTML;\n",
       "                element.style = originalStyle;\n",
       "            }, 2000);\n",
       "        })\n",
       "        .catch(err => {\n",
       "            console.error('Failed to copy:', err);\n",
       "            element.style.color = 'red';\n",
       "            element.innerHTML = \"Failed!\";\n",
       "            setTimeout(() => {\n",
       "                element.innerHTML = originalHTML;\n",
       "                element.style = originalStyle;\n",
       "            }, 2000);\n",
       "        });\n",
       "    return false;\n",
       "}\n",
       "\n",
       "document.querySelectorAll('.copy-paste-icon').forEach(function(element) {\n",
       "    const toggleableContent = element.closest('.sk-toggleable__content');\n",
       "    const paramPrefix = toggleableContent ? toggleableContent.dataset.paramPrefix : '';\n",
       "    const paramName = element.parentElement.nextElementSibling\n",
       "        .textContent.trim().split(' ')[0];\n",
       "    const fullParamName = paramPrefix ? `${paramPrefix}${paramName}` : paramName;\n",
       "\n",
       "    element.setAttribute('title', fullParamName);\n",
       "});\n",
       "\n",
       "\n",
       "/**\n",
       " * Adapted from Skrub\n",
       " * https://github.com/skrub-data/skrub/blob/403466d1d5d4dc76a7ef569b3f8228db59a31dc3/skrub/_reporting/_data/templates/report.js#L789\n",
       " * @returns \"light\" or \"dark\"\n",
       " */\n",
       "function detectTheme(element) {\n",
       "    const body = document.querySelector('body');\n",
       "\n",
       "    // Check VSCode theme\n",
       "    const themeKindAttr = body.getAttribute('data-vscode-theme-kind');\n",
       "    const themeNameAttr = body.getAttribute('data-vscode-theme-name');\n",
       "\n",
       "    if (themeKindAttr && themeNameAttr) {\n",
       "        const themeKind = themeKindAttr.toLowerCase();\n",
       "        const themeName = themeNameAttr.toLowerCase();\n",
       "\n",
       "        if (themeKind.includes(\"dark\") || themeName.includes(\"dark\")) {\n",
       "            return \"dark\";\n",
       "        }\n",
       "        if (themeKind.includes(\"light\") || themeName.includes(\"light\")) {\n",
       "            return \"light\";\n",
       "        }\n",
       "    }\n",
       "\n",
       "    // Check Jupyter theme\n",
       "    if (body.getAttribute('data-jp-theme-light') === 'false') {\n",
       "        return 'dark';\n",
       "    } else if (body.getAttribute('data-jp-theme-light') === 'true') {\n",
       "        return 'light';\n",
       "    }\n",
       "\n",
       "    // Guess based on a parent element's color\n",
       "    const color = window.getComputedStyle(element.parentNode, null).getPropertyValue('color');\n",
       "    const match = color.match(/^rgb\\s*\\(\\s*(\\d+)\\s*,\\s*(\\d+)\\s*,\\s*(\\d+)\\s*\\)\\s*$/i);\n",
       "    if (match) {\n",
       "        const [r, g, b] = [\n",
       "            parseFloat(match[1]),\n",
       "            parseFloat(match[2]),\n",
       "            parseFloat(match[3])\n",
       "        ];\n",
       "\n",
       "        // https://en.wikipedia.org/wiki/HSL_and_HSV#Lightness\n",
       "        const luma = 0.299 * r + 0.587 * g + 0.114 * b;\n",
       "\n",
       "        if (luma > 180) {\n",
       "            // If the text is very bright we have a dark theme\n",
       "            return 'dark';\n",
       "        }\n",
       "        if (luma < 75) {\n",
       "            // If the text is very dark we have a light theme\n",
       "            return 'light';\n",
       "        }\n",
       "        // Otherwise fall back to the next heuristic.\n",
       "    }\n",
       "\n",
       "    // Fallback to system preference\n",
       "    return window.matchMedia('(prefers-color-scheme: dark)').matches ? 'dark' : 'light';\n",
       "}\n",
       "\n",
       "\n",
       "function forceTheme(elementId) {\n",
       "    const estimatorElement = document.querySelector(`#${elementId}`);\n",
       "    if (estimatorElement === null) {\n",
       "        console.error(`Element with id ${elementId} not found.`);\n",
       "    } else {\n",
       "        const theme = detectTheme(estimatorElement);\n",
       "        estimatorElement.classList.add(theme);\n",
       "    }\n",
       "}\n",
       "\n",
       "forceTheme('sk-container-id-2');</script></body>"
      ],
      "text/plain": [
       "XGBRegressor(base_score=np.float64(0.013271975120564444), booster=None,\n",
       "             callbacks=None, colsample_bylevel=None, colsample_bynode=None,\n",
       "             colsample_bytree=0.5, device=None, early_stopping_rounds=20,\n",
       "             enable_categorical=False, eval_metric=None, feature_types=None,\n",
       "             feature_weights=None, gamma=None, grow_policy=None,\n",
       "             importance_type=None, interaction_constraints=None,\n",
       "             learning_rate=0.02, max_bin=None, max_cat_threshold=None,\n",
       "             max_cat_to_onehot=None, max_delta_step=None, max_depth=None,\n",
       "             max_leaves=None, min_child_weight=None, missing=nan,\n",
       "             monotone_constraints=None, multi_strategy=None, n_estimators=5000,\n",
       "             n_jobs=None, num_parallel_tree=None, ...)"
      ]
     },
     "execution_count": 4,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "model_depth3 = xgboost.XGBRegressor(\n",
    "    learning_rate=0.02,\n",
    "    subsample=0.2,\n",
    "    colsample_bytree=0.5,\n",
    "    n_estimators=5000,\n",
    "    base_score=y_strain.mean(),\n",
    "    early_stopping_rounds=20,\n",
    ")\n",
    "model_depth3.fit(\n",
    "    X_strain,\n",
    "    y_strain,\n",
    "    eval_set=[(X_valid, y_valid)],\n",
    "    verbose=500,\n",
    ")"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "7cdc8c89c7104fffa095e18ddfef8986",
   "metadata": {},
   "source": [
    "## Explain the depth 1 model"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "id": "b118ea5561624da68c537baed56e602f",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2026-04-09T21:21:10.073104Z",
     "iopub.status.busy": "2026-04-09T21:21:10.072926Z",
     "iopub.status.idle": "2026-04-09T21:21:11.981844Z",
     "shell.execute_reply": "2026-04-09T21:21:11.981055Z"
    }
   },
   "outputs": [],
   "source": [
    "explainer_depth1 = shap.TreeExplainer(model_depth1)\n",
    "shap_values_depth1 = explainer_depth1(X_test)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "938c804e27f84196a10c8828c723f798",
   "metadata": {},
   "source": [
    "### Bar plot shows the global importance of each feature"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "id": "504fb2a444614c0babb325280ed9130a",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2026-04-09T21:21:11.984340Z",
     "iopub.status.busy": "2026-04-09T21:21:11.984143Z",
     "iopub.status.idle": "2026-04-09T21:21:12.145815Z",
     "shell.execute_reply": "2026-04-09T21:21:12.145233Z"
    }
   },
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 800x650 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "shap.plots.bar(shap_values_depth1)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "59bbdb311c014d738909a11f9e486628",
   "metadata": {},
   "source": [
    "### Beeswarm plot shows the global importance of each feature and the distribution of effect sizes"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "id": "b43b363d81ae4b689946ece5c682cd59",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2026-04-09T21:21:12.147407Z",
     "iopub.status.busy": "2026-04-09T21:21:12.147266Z",
     "iopub.status.idle": "2026-04-09T21:21:12.429176Z",
     "shell.execute_reply": "2026-04-09T21:21:12.428632Z"
    }
   },
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 800x550 with 2 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "shap.plots.beeswarm(shap_values_depth1)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "8a65eabff63a45729fe45fb5ade58bdc",
   "metadata": {},
   "source": [
    "### The scatter plot for the top feature shows that XGBoost captured most the linear relationship\n",
    "\n",
    "It is important to note that XGBoost (and other gradient boosted tree models) is biased towards flat regions, which can be seen below by the flattened tails of the linear function."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "id": "c3933fab20d04ec698c2621248eb3be0",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2026-04-09T21:21:12.431114Z",
     "iopub.status.busy": "2026-04-09T21:21:12.430983Z",
     "iopub.status.idle": "2026-04-09T21:21:12.513917Z",
     "shell.execute_reply": "2026-04-09T21:21:12.513338Z"
    }
   },
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAjIAAAHFCAYAAADosxNlAAAAOnRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjEwLjgsIGh0dHBzOi8vbWF0cGxvdGxpYi5vcmcvwVt1zgAAAAlwSFlzAAAPYQAAD2EBqD+naQAAR6pJREFUeJzt3Qd4VFX6x/E3PaQQepciQUBYkaJggywgKk1QLCs2pK4FsetfFFlhXay44koHBRQVpYpK0Sw2kCIgTZoovRPS6/yf97gT02cmmUnmznw/zzOG3LkzObkTM7+c855zAmw2m00AAAAsKLCiGwAAAFBaBBkAAGBZBBkAAGBZBBkAAGBZBBkAAGBZBBkAAGBZBBkAAGBZBBkAAGBZBBkP0XUGk5KSzEcAAOAZBBkPSU5Olri4OPMRAAB4BkHGgZSUFOnZs6d06NBBduzY4aGXAQAAlAZBxoHp06dLdnZ2qS4uAADwLIJMCQ4cOCAff/yxDBs2zMMvAwAAKA2CTAlefvllufnmm6VRo0alurgAAMCzCDLFWLVqlezbt0+GDBni4ZcAAACUVnCpH+nD0tLS5I033pD7779foqKinHpMRkaGudkxWwkAAM8jyBRhxowZUr16denbt6/TF3LWrFkybdo0d742AADAAYJMAUePHpW5c+fKK6+8Yha0U6mpqblTsfUWERFR6EIOGjRIBg4cmK9HplevXo6uPwAAKAOCTAGHDx+WzMxMGTVqVKGLNWLECGndurXMnj270H2hoaHmBgAAyg9BpoDmzZvL5MmT8x3bvXu3vP766/LMM89Iq1atyvHlAQAAJSHIFBAdHW1W8S1Ky5YtpUWLFiVeUAAAUH6Yfg0AACyLHhknaA/Nhg0bPP9qAAAAl9AjAwCAm+w9nSmvf5sgH25JltMp+ffp08+/PZBe6HjBx09el2Q+lnS+M89VHryhHfTIAADgBiOXnpHFO9JyPw/4IkHGXxcjAy+NlHmbk+WFVQmSkS0SGiTyQvc/jhf3+JfiRQIDRHJshc935rnKg7e0gx4ZAADKSHtQ8oYYZRORMSsSzH32N3ylH8euSsjXi1HU4zXEFDxfb46eqzx4SzsUQQYAgDJatTe9yOOZNpHVe9Nz3/Dt0rNFdp3Mcvj4gufvPJHl8LnKg7e0QxFkAAAoo+6xYUUeDwkQ6RYbZoZe8goLEmlZM9jh4wue37JWsMPnKg/e0g5FkAEAoIxiq4fIjReHF3qDHdsjxtyn9SP6Rq/045juMVItIqjkxwcUPr96RJDD5yoP3tIOFWCz2f43Cgd30n2a4uLiJD4+3ukdtAEA1qa1Lkt3pkr9ykHSLTbcvOHbaf2IDr20qBmc73jBx3+1L126Ng2TqpUCiz3fmecqD97QDoKMhxBkAADwPIaWAACAZRFkAACAZRFkAACAZRFkAACAZRFkAACAZRFkAACAZRFkAACAZRFkAACAZRFkAACAZRFkAACAZRFkAACAZRFkAACAZRFkAACAZRFkAACAZQVXdAMAAPAWp1OyZeeJLAkPFlnza7pUqRQodaKC5KKawXIsMUfqRAeajy1rBcvZ1BxZtTdduseGSWz1kIpuut8iyAAAICLzNifLC6sSJCPb8eUIDBDJsf3x75fiRW68OFz+3aca17ECMLQEAPB72hPjbIhR9hBjt3hHmuw9nen317EiEGQAAH5Ph5OcDTHF+Wpfut9fx4rA0BIAwG+s3psqc35KkbvaRkijqsEyf0uKpGblSNOqwRISKJKZU/rn7tY0zJ1NhZMIMgAAv3DVO8fk0Pk/ksrX+8vWe5K3RkZpjUxTCn4rBEEGAOAXPTH2EONKWHns6igTUJrVCJbjSTlSOyrQfGxR849ZSzqc1LUps5YqEkEGAODzdDjJVdrj0rZ+mFzV6I8ho9jqku9j9Yggpl17AYp9AQA+T2tiXBUWJNKyJn/vezuCDADA53WLrSQNKpf8lhegwxQBf4aYMd1jpFpEUPk0EKVG1Czg22+/lffee0/2798vycnJUqtWLenSpYsMGzZMoqKiSn+lAQAV6ru/1zG1MvM2p8jAS/+YtfTR1hRJybRJm7oh0rVpuDlv18ksUwOjQ0fwfgE2m63Asj7+bfny5bJnzx5p3bq1xMTEyL59+2Tq1KnSokULefvtt51+nqSkJImLi5P4+HgCEAAAHkKPTAE9e/bM93mHDh0kNDRUxo8fLydPnpSaNWt66rUAAAAuokbGCdozozIzWX4aAABvQo9MMbKzsyUrK0t+/fVXmT59unTu3Fnq1atXvq8OAAAoEUGmGH369JETJ06Yf1955ZVmaKkkGRkZ5manhcIAAMCzKPYthhb8pqammtlLM2bMkPr165ti36CgoqvYp0yZItOmTSt0nGJfAAA8hyDjhN27d8sdd9wh//rXv6R79+5O98j06tWLIAMAgAcxtOSEZs2aSXBwsBw6dKjYc3Rmk94AAED5YdaSE7Zt22YKf3V4CQAAeA96ZAp44oknpGXLlqYXJiwszAwrzZkzx3yuC9wBAADvQZApoFWrVrJixQp59913JScnR+rWrSv9+/eXO++8U0JCQirmVQIAAEWi2NdD2KIAACrO6ZRsWbw9Rb7an2a2g+x6YZjc2CqC/ZN8ED0yAACfMm9zsjz7ZYLk3UjwmwMZ8o+vEmX8dTEy8NLICmwd3I1iXwCAT/XEPL8if4ix02PPr0ww58B3EGQAAD5j54ksySoqxfxPVo7IrpNZ5dkkeBhDSwAAS9EeldV70+RQQrY0qRYkK/akSVCASEylQAkJDCjxscGBIi1r8tbnS3g1AQCWqn8ZvSJBckrodSmORpx/XBsj1SKK3moG1kSQAQBYpidmzErXQ0yHesHSu2Ul6Xsxs5Z8EUEGAOC19p7OlPlbk+VsSo4cSsiSzBzXn+O65hEyqEOUJ5oHL0CQAQB4pZFLz8jiHboOTNl0axrmlvbAOzFrCQDglT0x7ggxN14cLk2rsyq7L6NHBgDgdVbtTXf63I4XhEjtyECJqRQklUMD5JfTWVIvOkjuaR8psYQYn0eQAQB4XW/MwXMZTp2r848m96vGTCQ/RpABAFi2LubF65hO7e8IMgAAy9TFXFwrUIIDAqRHs3C5o20Um0CCIAMAsE5dTP9WUTLscqZS40/MWgIAeIXusY6nSTOVGgURZAAAXkFnGOl06eIwlRpFoUYGAFDuVu9Nlbe+T5TzadmSnCHSrHqw7D+XJRlZNmlTO1ia1wqWyJAA2Xo0U5rWCJbhHaOZSo0iEWQAAOXqqneOyaHz+fcaOJacmfvvkylZsuV4lrSpGyxL7q7Fq4MSMbQEACjXnpiCIaY4W45myYZDzi+MB/9EkAEAlJs5P6W4dP6n21M91hb4BoIMAKDc3NU2wqXzB7Su5LG2wDcQZAAA5aZbbCVpUNm5tx6tkWlXn52rUTKKfQEAHjVnU6K8+V2iBAaIdG0aLjc0ryRZOTbZeDBdUrJyJElnLVULll911lK2zexW/fg1laVDA0IMHCPIAAA8psVrRyQ168/PP9j65xYEGmzG9YiRgZdG8gqg1BhaAgB4rCcmb4gpKMcmMmZVgpxOyeYVQKkRZAAAHjFzo+MZSpnZIrtOlpB2AAcIMgAAj7ivveMZSiFBIi1rUuWA0uOnBwDgVrqI3fivE2T/mZJ7WrRGZmz3GKkWEcQrgFIjyAAA3KbveyfMirzFqVYpQAb8JUIuqhFsZjBVJ8SgjAgyAAC39cSUFGLUmVSbXNcsnKnVcBtqZAAAbrFgm3PbCbDtANyJIAMAcAtntxNg2wG4E0NLBaxatUqWL18uu3btkvPnz0vDhg3ltttuk759+0pAQIBbLz4A+BJdiVe3FShpeIltB+BuBJkC5s2bJ3Xr1pVRo0ZJ1apVZd26dTJ+/Hg5fvy4DBs2jJ9AACjBkrtrmVqZf8YnyL7TWWZmUlSoSJNqITLySrYdgPsF2Gw2mwee17LOnTsnVapUyXdMg8yKFSvk66+/lsBA50bjkpKSJC4uTuLj4yUqKspDrQUAwL9RI1NAwRCjmjdvLsnJyZKa6lwhGwAAKB8MLTlh8+bNUqtWLYmMZGMzAMi7l9Kk75MkJdMmaZki9WICpFZkkBxJzJERHSPlrnbRXCx4HEHGiRCjw0paM1OSjIwMc7PTHhwA8AV7T2fKw0tOy66TOdKkqkhadqAcTMgpdN6BczY5cO6PQt/RKxNl/NeJsuuxehXQYvgTgkwJtMD3mWeekQ4dOsjtt99e4oWcNWuWTJs2zd2vDwBUqP5zTsimI3/OQtpzRv9bOMQURXe+1l4bembgSRT7FiMxMVGGDBliplxPnz7dYcFuUT0yvXr1otgXgGW1euOIJP35a61UYqsFyeqhtd3VJKAQemSKkJaWZoaSdOaR9rQ4M+soNDTU3ADAFyzcnlzmEKOGdHC8AzZQFgSZArKyssxw0oEDB8xQkRb5AoC/GbUsoczPUSlY5G9tKfiFZxFkCpgwYYJ88803pkdGh4d+/vnnfNOw6XUB4Ou0rsUVQXoL+HPW0rGkHBl2ObOWUD4IMgWsXbvWfJw4cWKhi7VkyRKpV48KfAC+bebGlBLvb1glUIIDcqRDg3AZ3jFaYquHlFvbgIIIMgUsXbq00EUCAH/SqmaQ7D+TXei4bjew8cHaUi1C+2AA70CQAQAYp1OypePbxyWzmNnV43rEEGLgddiiAAAg8zYnS7u3ig8xE3vHyMBLWd0c3ocgAwB+TntixqwseZbS0p1p5dYewBUEGQDwcztPZBXbE2N3d1vWg4F3IsgAgJ9rWavkcsm60YES17RSubUHcAXFvgDgZzYcSpexq87J7pPZkpFT8s5Jd7WNkHE9qpRj6wDXEGQAwI/0fe+EbDn65yaQJdEC3/6tKPCFd2NoCQD8aP8kZ0OMosAXVkCPDAD4gZFLz8jiHa7NPKLAF1ZAjwwA+Li9pzNdDjENKlPgC2ugRwYAfNyqvekunT/z5qrSLZZZSrAGemQAwMd1jw0r8f4AEQkNFBnaoZL89lQ9QgwshR4ZAPBx3aafLPa+NnWDZcndtcq1PYA70SMDAD6s0YQjxd43oFUYIQaWR5ABAD8MMerr/Rnl1hbAUwgyAOCHIUbd25aCXlgfQQYA/DDEaIHvyGvYegDWR5ABAB8ybvVZh+d0vTBEDjxVr1zaA3gas5YAwIfM35pa4v2bHqot1SOCyq09gKfRIwMAPuT2S4qve9E1Yggx8DUEGQDwIdM2pBYbYgBfxNASAFjcs1+ckblbit9L6c424eXaHsCre2RSUlJk7NixsmrVKs+0CADgdIDRGUolhRj16XbXNowEfLpHJiIiQlasWCFt2rTxTIsAACWavDZBXvpvstNX6dbW9MjAd5VqaOnCCy+Uo0ePur81AIAinU7Jlr8vPCXrDmW7fIXGXleNqwqfVapi37vvvlsWLFggv/32m/tbBADIZ97mZGn31vFShZi/X8bqvfBtpeqROXDggNSuXVtuv/12ufrqq6Vhw4YSHp6/6zIgIECGDBnirnYCgN/2xPzflwmleqyu3vt016pubxPgTQJsNpvN1Qdddtlljp84IEB+/PFH8VdJSUkSFxcn8fHxEhUVVdHNAeDD2w0U1dU+uEMlGd2NEAPfV6oemSVLlri/JQAAY+/pTHlg0SnZdcr5vzMjQ0S+GcGqvfA/pQoydevWdX9LAADSffpR2XPa5lLvy8cDq0uHBmFcPfilMi+Id+7cOTly5I+uz3r16kmVKuymCgClqYXRgl5n1Y4Ueen6qtItlmJe+LdSB5ndu3fLq6++Kps3b853/NJLL5UnnnhCmjVr5o72AYDPG/7pKfliT4ZLw0g/PsiWA0Cpg8zevXvNjKT09HTp0qWLWVdG7d+/X9asWWPumzlzpjRt2tSSV/ngwYMyZ84c2bZtm+zbt08aNWokH330UUU3C4APGbrgpKzYl+ny42LCRLaOIsQAZQoyU6ZMkeDgYJkxY0ahnhcNOcOGDZPJkyfLK6+8Ilak4eW7776TVq1aSU5OjrkBQEXORFJ3tY2QcT0YvgfKHGQ2bdokt9xyS5HDR7GxsTJgwAD55JNPxKo6d+5spk6rF154QXbs2FHRTQLgQz0xrurUIEj+07+GVI8I8kibAL8LMmlpaVK9evVi769Ro4Y5x6oCA0u14DEAOOTqcNJvTzGMBJSkVO/Y9evXl2+//bbY+/U+PQcA8KdWbzg/pPRMl0hCDOCpINOzZ0/54Ycf5NlnnzX1JNnZ2eam9TGjR4+WtWvXSu/evcWfZGRkmNV87bfkZOd3pgXg+2InHJEkJyYmBQb80QszolNMeTQL8M+hpbvuukt++eUXWbFihaxcudJsR6B0twO9de/eXe68807xJ7NmzZJp06ZVdDMAWLS4NyxQ5Llu0XJXu+hyaRPg10EmKChIXnrpJbnxxhvlv//9rxw+fNgc1+EkLZLt2LGj+JtBgwbJwIEDcz/XHplevXpVaJsAWCfE7H6CWhjAY0Fm7NixcvPNN0vr1q1zZy01adJEOnXqZG4QCQ0NNTcAsJuzKdGpi0GIATxcI7Ns2TI5dOhQ7ucjRoyQdevWleHLAoDvm7kxxeE5zEoCyqFHRvdPOn36dO7nWgfjy3TquH1W1tGjR80w0apVq8zn7du3l6pVq1ZwCwFYwX3tI2T0yuJ7ZQgxQDkFmUsuucRsOXDs2DGpXLmyOfbVV1+ZpfyLowXAulWBFZ05c0aefvrpfMfsn+uKxR06dKiglgGwkt/OZhU7tZpZSYB7BNic6F7R3a11hVvdIFJP15Di6GF6zo8//ij+Sqdga+FzfHy8REVFVXRzAJSzxhOOSFG/JRtUDpTv/l6H1wMozx6ZevXqydSpUyUzM9MMMfXp00cee+wxs2EkACC/cavPFhli1KHzObJ6b6p0i63EZQPKe/p1SEiI1KlTxyx2pzOY6tat6442AIBPmbkhtcT7521OIcgAFbmOzJgxY9z19QHA52Q7uP/uthHl1BLA97E7IgCU4wJ4USEicU0ZVgL8IshkZWWZuhwAsILYV0oOMbqZy/ZHWcEXqPChJXf65JNP5NNPP5WIiAi577775IorrjCbT06YMEF+/vlnycnJkdjYWHnggQfkqquuqujmAkCRFm5PlsyckkPMgacIMYBP9cjoPk3/+te/zHo0v//+uzz++OOydetWs3KwhhkNNboFgt7/6KOPyk8//VSRzQWAYs1ysIrvba3DuHqAr/XIzJ8/3+zZpIvt6VoruqfTk08+KdHR0TJ9+nSpXr26OU83pbz33ntl9uzZ0rZt24psMgAUaVD7CBm1LKHYqzOh1x+/zwB4WY+M9pboQnm6AJyr9u/fb9aksS8Yp7tH6zo1+tEeYuy7auumldu2bStrcwHAI55YXnyIYSsCwAuDzDfffCM33nijCRjDhg2TnTt35i7v369fv9y9iUqiexjZtzxQ9n/XqlWr0Lm1a9c25wOAt5m8NqHY+piHr2CqNeB1QWbDhg2mniUmJkaGDh2ab7uCatWqSYMGDWTFihUOn0cX19NaGLs9e/aYj7t37y50rh7TrwcA3ubttcXXx8zdXPLieAAqIMho/cpFF11kalZuueWWQvf/5S9/kV9++cXh81xzzTWycOFC+fDDD2X16tXy2muvmd6Y7du3y9q1a3PP096dxYsXUx8DwOucTsmWpIzi9567ty1rxgBeV+y7Y8cOGT58uAQGFp2DdBjo1KlTDp/nnnvuMTOXXn31VfN5UFCQmcWkNTM63bpSpUqmtyc1NdVsj6DTswHAW6Zb/2N1gpxx0OEy8poq5dUkwC+VKsjo2i6hoaHF3n/u3DkTPBypUqWKvP/++ybMaP3LZZddJhdccIG5b+LEifLuu+/KyZMn5cILL5RBgwaZXiAAqGit3zgiiRmOz1t4JzOVAK8MMjplWtd0KWpYyV4I7GzoCA8Pl+uuu67QcV1DRm8A4E3ipjoXYiJDRNrVZ+0YwCtrZHS2kta0LFq0KLfQNyAgQNLS0uSVV14xK/L279/f3W0FgArz2LLTZh+lX886d34qu6sA3tsjM2DAANmyZYuMHz/eDAFpiHn22WfNkJIOO+naMDfccIP7WwsA5WjDoXQZ9ulpOV2KiUddm1T4DjCAXyj1/2kvvviidO3aVZYvXy6//fab6Zlp1aqV9OrVS7p16+beVgJAOc9EuuLt45Jewt5Jjsy4tfB6WADcr0x/Mvz1r381NwDwhfBy+7zjsvtM2Z4nrnGwvHsbIQYoL/R9AhB/DzD93j0uv58v/XOwBQFgsSAzbdo0h+do3cyQIUNK8/QAUC7mbU6W//uy+D2SnEGIASwYZKZOnVpigNF6mbIEmYyMDFM4XLVqVafWowGA0vTElCXEDO1QSUZ3q8qFB6wYZJYsWVLoWHZ2thw6dMgscKc7Yb/wwgsuP++uXbvMLCjdTVtnP7399ttmkTzdiFJnRd17773SsWPH0jQZAPLZeSKrVOtV/OPaaLmrXTRXE7DyOjJ169YtdNONIjt16iRvvvmm2bpg6dKlLj2n7s2kPTgahnTmU166EWV6erosW7asNM0FgEJa1nL+77g+zUPNENKvT9UjxAC+EGRKokNKOv36s88+c+lxkydPlpo1a8pHH30kDz30UL4dtZX2zOhmkgDgDh9vTXJ4TmzVP2pgJvWrwUUH/GnWUmZmpiQkuDb2rMNJOnQUERFhamQKqlOnjlMbUQKAM176b3Kx91HAC/hxj4zujD1//nxp3LixS4/T8KK7XhdHN5UEAHeYvLZsM5UAWLxHRvdaKor2wqSkpEhQUJCMHj3apeesX7++KfYtzvr1681mlQDgyfVirryA5bUAKynV/7G1a9c2tTB56efNmzeXRo0amQ0j69Wr59JzXn/99TJjxgzp3r27eR77c6q5c+fKDz/8II899lhpmgvAz035MVH++XWiU+e+3a+6x9sDwH0CbAWraiuI1tU8+OCD8tNPP5lhqQMHDkhsbKycPXtWTp8+baZd22dEWYFOQY+Li5P4+PgSh8wAeE+IuS42TKbeTJABrMRrUoEufKfrxjz88MMSFhYmoaGhZjPKKlWqyMiRI+WNN96wTIgB4D3DSc6GGPXz8UyPtgeA+3nFYLAW+m7btk1q1KghAwcONDcAKKt2bx136fzB7Stx0QFfDDK6hkvBmhhH9Px169Y5da72tPz973+XRx55RBo2bCgVTYe1Xn75Zdm6datERkZKz5495f7772e7BMBCnv3CtW2sQwJFhnSM8Vh7AFRgkNGVdl0NMi41IjjY9MZ4Q7nO+fPnZcSIESZQvfLKK3LixAkzrJWWliZPPfVURTcPgJM+2Z7m1Hn6m+3pLpEyohMhBvDZIFOafZNcpasBr1y5Um677bYKrYX55JNPzJo1GmJiYmJy95GaMGGC3HfffWb1YQDeL7WErZSei4uk9wXwEV5TPduvXz/T6/HAAw/ImjVrzPDOsWPHCt087fvvv5fLL788N8Soa6+91mxiuXbtWo9/fQCls/d0plzzzhFpNOGPW0k9MAwhAb7DK4p9lfbE6PCVDi9t3Lix2PN+/PFHj7ZDA1Tfvn3zHYuOjjZDX3ofgIr1+ppz8uYPKaV+/OwBVd3aHgAWDTK6N9Ls2bPNbKPExMRC9S2uFPsq3fnak3U4rtTIaHApSI/pfSXNvMq7RxRbKgDu1+TlI5JTxlK6uKbMTALE34PMpk2bzCweXeitdevW8t1335mZTbo9ge5QrQvZtWjRwqXnHD58uFjZrFmzZNq0aRXdDMCne2LKGmLYDBLwPaUKMjNnzjRDLXPmzDG9KFpDMmjQIBNmtI5EZ/dYdYZP5cqVzaq8BWmvk95XHP3+865/oz0yOtsLgHtM/rH0w0mqc8MgXgrAB5UqyGivi75pV61a1WwUqbQYVnXq1MmsuzJ58mRzc6WXxxnt2rUTT7Jvj5CXBptTp06VuKO3rkSsNwCekZNdtsfP+VttdzUFgNWDjNaC2Kch29+8dVjJ7qKLLpLly5e7PLTkTI2Mp4t9r7zySjNMpD0w9lqZVatWmSnhGtIAVIx7O1SSaRtSXX7clRcEyQd3EGIAX1WqIKPDSrpQnKpUqZJ5w9+3b5/89a9/Ncf0Pl3kzhVjxowpdCwrK0sOHz4sS5cuNbtp33TTTeJpN998s3z44Ydmp21dN0a/F92sUr82a8gAFWd0t6oyfUOqOCqT6dM8VCb1q1FOrQJgySBz8cUXy5YtW3I/152p33//falTp46ZvfTRRx9Jq1atXHrO3r17F3vfXXfdJXfeeaeUB62Deeedd8yCeBpmdIsCXeNGi5sBVKwDT9WTcavPyodbU+W2SyqZcAPAvwXYSrEvgBb0Llu2TEaPHi3h4eFy6NAhGTp0qKkjUdWrV5dJkyaZ2UvuMmPGDPnyyy9NSLICrauJi4uT+Ph4M7sLAAB4SY+M1orkrRdp0KCBfPrpp6Z+JSgoSC699FK3v3lrT4kOMwEAALh9ZV+tlenSpYt4Qnp6uike1p4eAP5rzqZEmbwuWVrVCpEn4ypLbPWQim4SACsGGZ163adPH7n++uulSpUqbmnI2LFjizyuq+n+/PPPcvbsWXn44Yfd8rUAWE+L147kbgR56Hy6fLn3pNx4cbj8u0+1im4aAKvVyOg6MSdPnjQzk3S6si781rlzZ5dnKuWli+kVN6TUqFEjufXWW01wsgpqZAD39sSMXplY5H2rh9SkZwbwY6VKHp999pnZR0k//ve//5VvvvnGTMHu0aOHmX3k6owltX79+tI0BYAfKC7EqK/2pRNkAD9Wqh6ZvFJTU2X16tUm1OjqvPp0DRs2NIHm3nvvdfp5jh07ZoapdBZUUdLS0uTcuXNmircV0CMDlM09H56Q+AP/G0sqwVdDakpTamUAvxXojiJfDS269sqSJUvk73//u5mGrZ+7om/fvmaqcnHWrFljzgHg+xpNOOJUiIkKEUIM4OfcNmtJ15LRXpnPP//cbJjoar2Mo44h3cvJmS0MAFg/xDhr9F//2EYEgP8KLuvwyYoVK0yA0ZlFGkaaNWsmo0aNkhtuuMG9K3oeOMDCcoCPevaLM/Lh1jTJdGGgW/+s+Vtbggzg70oVZLS4V8OLftQNJKtVqya33367GWLSDSOdpasD681u5syZsmjRoiKnYOteTrpSLgD/7YEpuF0BAJQqyDz66KNm1+trrrnGhJcrrrjCrOjrKt1h+siRP36J6bCRrhWjRb156XGtw9H6GPY7AnyvJ6Y0fiPEACjLrKUFCxbIddddZ6Zcu4uuI/Piiy9aaq2YkjBrCXB/b8y4a6PlrnYMJwEoY4/MgAEDxN1YRwbwL66EmH/1iKYeBoBnpl8DgKu6TXU+xFQKpqgXQDlMv3bXFO558+bJ9u3bTYFvUaNeixcvrpC2AXCPxhOOiDPj2ZVDRZ7swlASAIsEmb1798rgwYMlMzPT7K10+PBhufDCCyUhIUFOnz4tDRo0kFq1alV0MwGUwbjVZ50KMRTzArBckJk8ebKEhITIu+++a7YquPbaa+Xxxx83RcALFy6Ut99+W1577bWKbiaAUnhs2WlZsD3dqbVhmFYNwJI1Mps3b5b+/ftL48aNc4/Zh5b0uO6y/dZbb1VgCwGUtqjXmRBTPZwQA8DDPTJZWVlmt+uDBw+aXhNdoE4/ukNKSooZPlLaM2PfkNKuTZs2plcGgHUMmHPM6XM3PcwCdwA8GGS0+Hb48OFmhV3tKdGF6rSHZNKkSdKyZUspK10dWGthVGRkpFkE7/fff8+3eF52dnaZvw4A7xlKylvYCwAeDTIzZswwBblXX321WclXQ8Ynn3wi48ePl7lz50pZ6dYGO3fuzP28Xbt2Mn/+fGnVqpXZMPKjjz5yafsDANbZcmBgm0oeaQsA3+d0kNF9lTTAvPHGG7nH6tatK2+++aYcP35cateuXaaG6Iq+H3/8sdmiIDw8XEaMGGF6gPSjCgsLkwceeKBMXwOA94UYLfB9umtVj7QHgO9zOshoWLntttvyHevcubNMnDhRjh49WuYg06NHD3Oza9GihXz44YcSHx8vgYGBptjXXkMDwPohRndnu69DJRndjRADoByCjO5yHRMTk++Yfa8lXfvFE+rUqWN21QbgOyGGNWIAeN06Mlr46y46U2nr1q1y5swZufzyy6V69epue24A7keIAWCZIKNFvV9++WXu5zqLSEPMf/7zn0K9NXr89ddfd3lXbZ0FlZycbB6v0601yGio6d27tzzxxBNmTRkA3uHZL844dd7DV0TIo53ds1QDAJQ6yPzyyy/mVtDPP/9c5l6a1atXy4QJE6RLly5yzTXXyLhx4/JNzdZCY62XIcgA3mHk0jOyeEeawzqY/U+xPgwALwgy69ev92AzRObMmSMdOnSQV199Vc6dO5cvyChdq2bRokUebQMA5+w9nekwxOgvl32EGAD+skWBrlGjKwUXp0aNGnL27NlybROAwgHmjg9OSLfpJx1eGkIMAEsGGV2dVzd+vOWWW1x6XFBQUO7eSkU5deqUWV8GQMUNJWmA+e73LIfnLryTIn0AFpq1pCvv6oJ5ixcvlu+//94UAUdERLj0HM2aNZMffvihyOnW+vyrVq0yq/wC8M6hJLs2dYOlXf0wj7cJAMocZA4cOCBLliyR5cuXm5lFuq7MDTfcIN26dZOOHTu69Fy33nqrPPvss/LOO+9Ir169cgOMfg2dvbR//3556KGHeNWACrBqr3P7Jl15QbB8cEctj7cHAEodZHSdlxUrVpgAo7OVdEhId6bWIKNBpGvXrlIauqqv1snMnDlTZs+ebY6NHDnSDDfpbdiwYXLVVVeJp61du1aWLl0q27Ztk8OHD5shsqeeesrjXxfwZt1jw+SleMfnvd2PISUAXhpkNm/ebMKLTpNOSUmR5s2by6OPPmr2SNKdqW+66aYyN+b+++83Qejzzz83PTEaYBo2bCg9e/aUiy++WMqDDm/t2bPHbFqpO34DEKeKe/95XYxUi9AJ1wDghUFm6NChZj0XXcdFF6eLjY3NvS8pKcnlLzx27Fi5+eabpXXr1ubzTZs2SZMmTcweS3qrKA8//LA88sgj5t8bNmyosHYA3uLi10refqB+lMjSQbWlOiEGgLfPWkpPTzehpTTBpaBly5bJoUOHcj/XXa7XrVsnFU03qAT83eq9qXLlO0fM9gPJJUxSqhwq8v0D9QgxALy/R+bjjz82C9LpsI8OMdWrV8/0zNgLc11VpUoVM1XbrqSp1wDKz1XvHJND53OcOndgm0oebw8AuCXING7cWEaNGiUPPvigrFmzxky1njZtmrnpMJNuSeBKGLnkkktMYe+xY8ekcuXK5thXX30lBw8eLPYx+jWGDBki3kh3B9ebne4XBVixJ8bZEKObkDzdtarH2wQAJQmwlaErRBep094ZneWjw0ShoaFmTyQt2O3cubNERUUV+9gjR47ICy+8YIqItQnOBCE958cff3SpjToMpu10pH79+hISEpLvWJ8+feTqq692atbSlClTTKgrSPeHKuk6AN7k3o9Py9f7nZtq/RvbDwCwepDJa+PGjaaXRntVtJZGQ4EujudIZmamGWLS0PDYY4+ZTSNLUrduXZfapcNhBfdtKm7nbe11Km2QKapHRofdCDKwkjmbEmX0ykSH5/VvESoTb6xRLm0CAI+v7Kvat29vbk8++aR88cUXpqfGGRp46tSpY+ptdAaTq0HFkX79+pmbp2lvlN4Aqxr44Sn59sCfYbwkhBgAPhdk7HQYZcCAAebmijFjxri7KQCc3H6g58yTkp7juCamd/NQmdSPnhgAFgwyCQkJLj95TEyMWM3Ro0dl+/bt5t9paWlmdV/d50l17969glsHuM+GQ+mmJibRiU6YtnVDZNHdNbn8AKwbZPRNXIttnaXnesO6MK7SRfB0sT47rfOx1/qwQB78qQ4mr8EdXNsEFgC8Lsho4WreIKMFvStXrpROnTpJjRq+09WsBb56A3xRi9eOSGoJC9wVpUp4gPS5ONJTTQKA8gkyOlU6r3Pnzpkgc/fdd8tll11WtlYAKJeeGFdDzMTeMdK/FSEGgB8V+wLwTq4MJzWIFvnu/noebQ8AuANBBvADk9c6X6y/ekhNia2ef3FIAPBWBBnAD7z5reMtM4Z2qCSju7HlAABrIcgAPmrogpOyYl+m4/MIMAD8Oci4MiUbQPloNOGIU+c1qBxILwwA/wgyt99+e77Pc3JyTIh58cUXpVKlSoXO1/s++OAD97QSgEs9Mc56uks0VxaAfwQZ3QSxYO+L7pGke06mpKR4om0ASmGVE8NJ9v/5WR8GgN8EmaVLl3q2JQDconvTEKdqY/Y9xfRqANYXWNENAOBejkLMgFZh8hshBoCPcMuspaysLLPR4smTJ6VJkybStGlTdzwtAAdW702Vf8YnyMnkHIkIETlawpp3PZqGyLQBbPwIwE+DjG6Y+PXXX8vgwYOlWrVqucd1d+jHH39c9u3bl29fpjFjxri/tQByXTbpqJxItuV+npBW/MWpFCyEGAD+PbS0bNky+eGHH/KFGKU7Re/du1cuueQSueOOO0yPzGeffWbOB+AZt75/Il+IcaQmm1cD8Pcgo0NHutN1XgcOHJCffvpJ2rZtK9OnT5dRo0bJu+++KxdccIEJMwDcb+/pTFl30LXdH1/swYq9APw8yJw+fVoaNmxYaLhJp2T369cv91h4eLhcf/31smfPHve2FICxam+6S1ciIlgkrmnhtZ4AwK+CTEZGhoSFheU7tmPHDvOxXbt2+Y7Xrl1bkpKS3NVGAHl0j83//6EjOx9jmjUA3+V0kNHF7/bv35/v2ObNm6Vq1armvrzS0tIkOpoVQwFPuHnuqSKPB4lIZJ7yfaZZA/AHTs9a0joYrXu58cYbJTY21sxgOnjwoPTu3bvQuVr8W7Mm0zwBd9pwKF2e+fKsnEsrush3/UO1pXqExhkA8B9OB5l7771XPv/8czMzKSYmRhISEiQkJETuvPPOfOdlZ2fLmjVrpGvXrp5oL+CX+r53QrYcLbnAd9fJLLmqEUEGgH9xemipfv36MnXqVLnqqqtMkLnyyitlypQphRa/0wJgvb9Lly6eaC/glz0xjkKM7oLWsqZb1rcEAEsJsOmuj3A7LXaOi4uT+Ph4iYqK4gqj1J7+4px8sKXkjVkHd4iU57vFcJUB+B32WgK83IDWjqdOP3gFYRmAfyLIAF6uQ4MwaVO3+GGjf14XI9Uo8gXgpxhUB7zcuNVn5ec8NTKRISJiE+nRLFye6x7DTCUAfo0gA3ixxhOOaGbJJzlTJCRQZGLf/PueAYA/IsgAXraP0vBPT8reMyWfl5kjMnltgozoRIEvAP9GkAG8xPUzj8nOkzlOn//uplSCDAC/R5ABvECTl49IjosLIQxuz0aQAMCsJaCCXT7J9RCjNTJDOjKsBAD0yAAVaPXeVDme7NpjnukSyZASAPwPQQaoIPM2J8v/fZng0mN+e6qex9oDAFbE0BJQAU6nZMvzK0sOMSF5/tLo0TSEEAMARaBHpsDO3XPnzpVvv/1W9u/fL7oNVbNmzWTEiBHStm3boq4fUKoF7mZvTJWsEupidLXegZdGcnUBwAGCTB7p6ekye/Zs6d27t9xzzz0SGBgoCxcuNEFm0qRJctlllzm6noDLC9wV9FbvGOnbihADAM4gyOQRFhYmixcvlsqVK+ce69ixo9x2223y/vvvE2RQ5p4YRyGmblQAIQYAXECNTB5BQUH5Qoz9mA4vnTx50pXrChQyf2uqw6vyWm+2HQAAVxBkHMjKypKff/5ZmjRp4tKFBQq64gIt3y1eUIBIy5p0kgKAK/it6cB7771nemPuuOOOEs/LyMgwN7vkZBcXB4FPu+qdY3LofMnbD7zYI0aqRQSVW5sAwBf4fJBJSkqSU6dOOTyvfv36EhKS/y/mtWvXypQpU2TIkCHSsmXLEh8/a9YsmTZtWpnbC9/z7Bdnig0x4YEi/VuHyxNdYqQ6IQYAXBZg0znGPmzRokUybtw4h+ctWLBAGjdunPv5rl27ZPjw4RIXFydjx451+PiiemR69eol8fHxEhUVVYbvAFbW4rUjkppV/P1t64bIortrlmeTAMCn+HyPTL9+/czNFQcPHpSRI0fKJZdcIs8995xTjwkNDTU3YO/pTJn47XlZvTe9xBCjBneI4IIBQBn4fJBxlQ5DPfjgg1KnTh2ZMGGCBAdzieB4ld55PyXJh1uS5XCiOJxibVclPED6XMx6MQBQFrxL55GWlmZ6Ys6dOyePPfaY7Nu3L/c+rZ9p0aJFmS42fE9p9ktS/9clUoZ3YvdqACgrgkweZ86ckd27d5t/P/roo/kuVN26dWXp0qVlvuDwrZ6Y0aUIMTdeHE6IAQA3IcjkUa9ePdmwYYO7ri183M4TWVLyhOrCno2LlGEd6YkBAHdhQTyglFrWCnbpfyCtiSHEAIB70SMDlNLmIxlSI0LkRErJ5zWIFnm8S4z0ZyNIAHA7ggzg5pV6L6gsUisyWB64Mlq6xVbi+gKABxFkABet3pta4nYDY6+tSoABgHJCjQzgojk/lTyWNG+zg7EmAIDb0CMDuLBi7ztrz8uBs39uRVGUu9uyWi8AlBeCDOCEkUvPyOIdaQ7Pa1A5UOKaUhcDAOWFIAM40RPjKMTUjRYZ34PaGAAob9TIAA7M2pDs8BpdXCuMAl8AqAD0yADF1sMkypIdaZLhxPK91MUAQMUgyAClrIexoy4GACoOQQb4n4Xbk+WVNefl8HmbU9fkL7WC5JFrKjOkBAAViCADiEibN4/KuTTnAowKEpH3bqsh1SL0XwCAikKxL/zenE2JLoUY9eJ1MYQYAPAC9MjAr83bnCyjVyY6nfofvTpS7mgbJdXpiQEAr0CQgd86nZItz32Z4NS5466NlrvaRXu8TQAA1xBk4Ld2nsiSbCfO++d1MTLw0shyaBEAwFUEGfitlrWCTdFuUWGmVqTILa0jZPDl0QwjAYAXI8jAb4aR1h1Ml8Q0myRn5MhPRzLkYEK2xMWGyuq9+TeBbFM3WJbcXavC2goAcB5BBn5R0PvcigTJLmFiUq/mYRIeHCB3XBopHRqElWfzAABlQJCBz/fEjFlZcohRX+5Jlx8fqM0wEgBYDOvIwOcLejOd2CspK0dk18ms8mgSAMCNCDLw+YLeECd+yvWcljXpoAQAqyHIwKfpwnVjr42RYAc/6XoO2w0AgPXwJyh8nq4Bc/1F4fLjwQxJTM8xs5Y2Hc6QQ+ezpVvTcPnbpZHUxgCARRFk4Dc9Mzc0r5T7+aAOFdocAICbEGTge+vF/J4hEmCTsKAAWbAtVYICbFI3OlhuaxMhsdVDKrqJAAA3IsjAT9aLSZep65PlxovD5d99qpV/4wAAHkGQgSV7XVbvTZMtRzMlIiRAGlYJlLmbU2TXScc7Jy3ekSYjr8ykZwYAfARBBpbrdRm9IkFyHCxwV5Kv9qUTZADARzD9GpZbpbcsIUZ1a8oWBADgKwgy8LlVekuiNTJNKfgFAJ/B0FIB7733nnzxxRdy5MgRycrKkvr168tNN90kt956qwQEBFTMq4R8q/Q6G2bG94iW73/PlCARqVs5SG69hFlLAOBrCDIFJCYmSo8ePaRp06YSGhoq69evl1dffVWSk5Plvvvuq5hXCflW6XWmRkZ7Xu5sGy13tuXiAYAvC7DZbGWsOPB9o0ePlh07dsinn37q9GOSkpIkLi5O4uPjJSoqyqPt88dama/2/Tlr6YKYQPno51RJy8qRNnXCZESnKIp5AcBP0CPjhJiYGMnMzPT8qwGne2Zu+Uuk3PKXP4/d1S6aqwcAfoggUwytj0lPT5effvpJPvvsMxk6dGj5vjIAAMAhgkwRDh48KP3798/9fPDgwTJw4MASL2RGRoa52WlNDQAA8CyfDzJaq3Lq1CmH5+nspJCQP/bhqV27tpm9lJKSIps3b5bZs2dLYGCgDB8+vNjHz5o1S6ZNm+bWtgMAAD8v9l20aJGMGzfO4XkLFiyQxo0bF3nfBx98IBMnTjRDTDVq1HC6R6ZXr14U+wIA4EE+3yPTr18/cyuLli1bSnZ2thw9erTYIKNTtfUGAADKDyv7OkGHl3QxvHr16nn+FQEAAE7z+R4ZV+tpRo4cKT179pQGDRqYmUsbN26U+fPnm9V9q1evXtFNBAAAeRBk8tChoUaNGsm8efPkxIkTEh4ebgLNM888Y+pdAACAdyHIFAgyY8aMqbhXAwAAuIQaGQAAYFkEGQAAYFkEGQAAYFnUyMCju1Sv+z1Dl12Ui2qEyMZDGXI4MUs6Nw6XtCyR8GCRNQfSpH50sHRvFm4es+5guiSm2SQ6PMA85lhijjlv4Y4UOZNik8EdIqVDgzBeNQCAQZCBR8zbnCzPrUiQ7CLWjX7zuyL2ofoiQQIDRHIcrDO9/Jc0aVM3WJbcXct9jQUAWBZDS/BIT8yYlUWHmJI4CjF2W45myYZD6aVqGwDAtxBk4HY7T2RJZo5nL+yn21M9+wUAAJZAkIHbtawVLCEe/ska0LqSZ78AAMASCDJwu+oRQTL22hgJDnD+MXqq1sg4Q2tk2tWn4BcAQLEvPGTgpZFy/UXh8uPBDPN5sxrBsulwhhw+ny3XNA6T9GyRsCCRbw6kS/3KQdIt9o9ZS3p+YnqORIcFmsccT8ox5y3akSpnU3NkUHtmLQEA/hRgs9lcLMmEsxtQxsXFSXx8vERFRXHRAADwAIaWAACAZRFkAACAZRFkAACAZRFkAACAZRFkAACAZRFkAACAZRFkAACAZRFkAACAZRFkAACAZRFkAACAZRFkAACAZRFkAACAZRFkAACAZRFkAACAZRFkAACAZRFkAACAZRFkAACAZRFkAACAZRFkAACAZQVXdAMAwN2OHDni8Jx69epx4QEfQI8MAACwLHpkSrBz50655557JCwsTL755pvye1UAlLqnBYB/IcgUw2azycsvvyxVq1aVlJSU8n1VAD9U3iHFnV+PYSqg4jC0VIwlS5bIuXPnpG/fvuX7igAAAKfRI1OExMREmTRpkjz//POyY8cO8RanU7Jl54ksaVnrj5fN/u/qEUEuP97+mKKOOXqMo+dbdzBdEtNzJDo0SDo2DHW6fbAmhnsoLgYqEkGmCP/5z3+kZcuWcs0113hNkJm3OVleWJUgGdkiQQF/HMu2iYQGibzQPUYGXhrp9OPtj1EFj+V9nqIeY7+/uOd7bkWCaZedtvXFHo7bBwBAaRBkCvjll1/MsNK8efNcupAZGRnmZpecnCzuor0c9tCg8gYFPTZ2VYJcf1F4sT0fBR+vH19YmSD6NJk5RT9PUY+x36+Ke768bbO3Ve8rqX2AP2BKOOAZPh9kkpKS5NSpUw7Pq1+/vgQHB8uECRNkwIAB0rhxY5e+zqxZs2TatGniCTp8Yw8NRUnPFtl1MkuuahTk9OMzckp+nqIeY7/fZhOnni/vfSW1D+7FG6bvD9NRXAz4UZBZtWqVjBs3zuF5CxYsML0xBw4ckPHjx5s6GWXvZdHPQ0NDzVTsogwaNEgGDhyYr0emV69ebvketAZFh2+KCzNhQSItawa79PjQQMnXI1PweYp6jP1+fZwzz5f3vpLaB+9E7QsAKwiw6TxjGFOmTCmxV0XXlHnooYec7gmKi4uT+Ph4iYqKKvMV1poUHdrRXpFgnWtmE8my/REuxjhZI2N/vP0xquCxgjUyxd1f3PM9vzJBsvKEmeAAkX9QI1OuCCBQ9NrAXxBkCrwBFHwTWLZsmaxcuVLefPNNqVOnjjRo0KBCgozSuhUdomnxv94N+79dmbVU8DFFHXP0GEfP9+PBjD9mLYUFyuUXMGupvBFkoAgy8BcEGSd6aebOnevyyr6eCDKAMwgycBZhB76ABfEAAIBlEWQcGD58OPssAQDgpZhKAngBpkwDQOnQIwMAACyLHhkA8FP0BMIXEGQAi6zCymwkVAR3/twxSwqeQJABPIwAAgCeQ5ABikD4AABrIMgAAMoFNTnwBIIM/A69LYD3YgdwuIogA0vglxsAV38nUFzsHwgy8KlfNvS2APDE7wNCkfciyHiIzWYzH5OTk8v8XEePHnV4Tt26dcv1uZx5Hmft2bPHbc8FAJ5Qnr+nnP197g8iIyMlICCgxHMIMh6SkpJiPvbq1ctTXwIAAJ8WHx8vUVFRJZ4TYLN3HcCtcnJy5OTJkxIREeEwTXoD7TnS0PXZZ5+ZBOzr/On79afvVfH9+i5eW/97bSPpkak4gYGBUrt2bbEa/aFxlH59iT99v/70vSq+X9/Fa+u7Ikvxe4pNIwEAgGURZAAAgGURZGCEhobK0KFDzUd/4E/frz99r4rv13fx2vqu0DL8nqLYFwAAWBY9MgAAwLIIMgAAwLIIMgAAwLJY2RdFeu+99+SLL74we5VkZWVJ/fr15aabbpJbb73VEgv8OSs7O1vmzp0r3377rezfv99sLdGsWTMZMWKEtG3bVnzR2rVrZenSpbJt2zY5fPiw3HLLLfLUU0+J1R04cEBefvll2bp1q1mLomfPnnL//fdLSEiI+JqDBw/KnDlzzGu4b98+adSokXz00Ufii1atWiXLly+XXbt2yfnz56Vhw4Zy2223Sd++fX3qd5Gd/i7S37/6+0gXiatVq5Z06dJFhg0b5vPrQKWkpMiAAQPkxIkT5hpcfPHFTj2OIIMiJSYmSo8ePaRp06aminz9+vXy6quvmv+x7rvvPp+5aunp6TJ79mzp3bu33HPPPWYhw4ULF5ogM2nSJLnsssvE1/zwww9m35h27dqZNwZfoN+Hvmb6JvfKK6+YX4RvvPGGpKWl+URIK0jDy3fffSetWrUyq4jrzVfNmzfP7D00atQoqVq1qqxbt07Gjx8vx48fN2/uvkZ/lvV11bAWExNjXuupU6eaj2+//bb4sunTp5s/Ll2mWxQAznj22Wdt/fv396mLlZWVZUtISCh07Oabb7aNGjXK5ouys7Nz/927d2/bv/71L5vVzZw503b11Vfbzp07l3vsk08+sV1++eW2EydO2Hz5NRwzZoztlltusfmqs2fPFjo2btw4W+fOnfNdB1/26aef2tq3b++TP8t2v/76q/l/eMGCBeZ73b59u81Z1MjAafrXQWZmpk9dsaCgIKlcuXKhYzq8pHtl+SLtdfI133//vVx++eXmZ9Tu2muvNT0VOpTma3zxNSxOlSpVCh1r3ry56R1OTU0Vf2D/ufa137956bDwzTffbIZJXeU//zegVLQ+Rn9h6LitbuZ1++23+8X3/PPPP0uTJk0quilwoT6mcePG+Y5FR0dLjRo1zH3wLZs3bza1I768CaoOsejQt9YG6ZBL586dpV69euKrdVD79u2TIUOGlOrx1MigxILC/v37534+ePBgGThwoM9fMS0y096YO+64o6KbAhfqCjS4FKTHfKUOCH+GmBUrVpiaGV/Wp08fU+ulrrzySlMX5IvS0tJMPZsW5pe2mJkg4yeSkpLk1KlTDs/T2Un2WR66e7e+qWsluf7y0KJY7dIePny4+Nr3aqfDEFOmTDF/GbRs2VKsoCzfL2AlWuD7zDPPSIcOHXy+d/jNN980Q2c6e2nGjBnyyCOPmGJfHfr2JTNmzJDq1aubWWilRZDxE9p1N27cOIfnLViwILeLXmcr2ae/6S8O7cadOHGiGcfULntf+l6VduHqDJfrr7/e7PlhFaX9fn2J1jlpoCtq9l3BGihYk76WI0eONPUiWk/h63VCWqenLrnkEvN7WHuIv/76a+nevbv4iqNHj5rlL3Smof3/X3vdk/4BrbeIiAiHz0OQ8RP9+vUzt7LQHgodt9UfPm8OMqX5XnUYTX9J6i+N5557TvzttbU6DWgFa2HsPVW+Gt78iQ4/6FCSvqazZs3y+fVUigo1wcHBcujQIfElhw8fNgXMRQ0T6nIKrVu3NiMBjhBk4DQdXtIFqHyt4Ezf7B588EGpU6eOTJgwwfzCgLVoDYG+welf7fZaGe2p0r/aO3XqVNHNQxmL73U4SYPqtGnTTJGvv9GFD+0Lk/qS5s2by+TJk/Md2717t7z++uvmNdf1dJzBb2wUon/1aO+ErozaoEED8z/Qxo0bZf78+WZ1Xx3P9KW/9PR7PXfunDz22GOmct5O60latGghvkZ71LZv3577/etfRfqmr6zaba3DnR9++KF5DXXBRi2S1BoD/XmtWbOm+Bp93XQmof311JmF9tewffv2ZuE4X6F/XHzzzTfmr3b9PnVGYd43Qh0C9yVPPPGE6f3WXpiwsDDzxq6rOOvncXFx4kuio6NN2UJR9Bo4+/s3QBeTcXPbYHEZGRny0ksvmR4YfUMIDw83gUbfLHr16uVTxWa6BUNxRWa6mqgu5e9r9HsaO3Zskfdt2LBBrOrXX381Y+1btmwx9Vz6s+qrWxSU9HOrf+EW9+Zg1dk7GtaKsmTJEp/rIdahFJ2VpX9g6DpI+nuoa9eucuedd/rFkNqGDRvMsJIrWxQQZAAAgGX5dtk3AADwaQQZAABgWQQZAABgWQQZAABgWQQZAABgWQQZAABgWQQZAABgWQQZAB6ji3rparu6YrAu0vbCCy9wtQG4FVsUAPAYXUF4z549ZtsA3dpCV4j2pPfff98se66rwXo73Rbj3XfflTVr1sixY8fMqq1NmjSR22+/3eeWogc8iZV9AXhsq4urrrpKbr31VrN/THnQAKNLuk+dOlW8fa+kO+64Q44fP252Ltd9dBISEmTZsmVmq4Wnn35aBgwYUNHNBCyBHhkAHnHmzBnRrdwqV67sE1dYN0/Nzs42G/mVVXx8vPz+++9m2O1vf/tb7nHd5FI3a/30008JMoCTqJEB4HZaC9O7d2/z72nTppn6GL3l3ZRSN8YbPHiwdO7c2fTc3HPPPbk7OOel5z3yyCNmE8grrrhCunXrZgKADlnlpc+vmwtu2rQp9+vpTTdYtN9fVI2ObqJZsG1Tpkwxx3Q39Ndff92EiyuvvDJ352XtbZo5c6bpbdLjOhSkbdy1a5dT10d3cVYFd+bW4aVKlSqZjVoBOIceGQBupz0LF110kQkBf/3rX81NaQ2I+s9//mOCgIYA3ek2MDBQvv76azOk8uSTT5qAYPfRRx9JTEyM9O/fX2rUqCGHDh2ShQsXmhA0d+5cadiwoTnvH//4h/l6VapUMTU5dlWrVi319/Hcc8+ZHpiBAwdKQECA+fraM/PQQw/J1q1bTcDRtiYlJeW2SYObo117L7vsMrOL/KRJk0xo0aGlxMREmTdvnvmYt/0AHLABgAccPnzY1r59e9vkyZPzHd+5c6c5PmnSpEKPefTRR22dO3e2JSUl5R5LSUkpdN7+/fttnTp1sr300kv5jvfu3ds2dOjQItujX3PMmDGFji9ZssTct379+txj2mY9ps+VmZmZ7/y5c+ea+77//vt8xxMTE209e/Ys9usXtHr1anO+Ppf91qNHD9tPP/3k1OMB/IGhJQDl6vPPPze9GzpUpDN38t50mEmHXexDOEqHWv73R5fp+dDztJelUaNGsm3bNo+2VQtyg4ODC7W/cePG0rJly3xt156ajh07ypYtW0wxryM6uyo2NlaGDRsmr776qjz11FOmd0aHzXbv3u3B7wrwLQwtAShXOitHQ0lJs3JOnz6d+2+tO5k8ebJs3LhRUlNT851Xv359j7bVPmxVsP3p6elmbZziaLCpU6dOsff/8MMP8vDDD8vEiRPN8JqdDsHpdZkwYYLMmDHDDd8B4PsIMgDKnfbI/Pvf/za1MUVp2rSp+ajrq2iPRWRkpKk/0Z4Q7bXQx7/22muFgk1p6Eyk4hRXdKs9KVrcWxxHdTm6foz2NOUNMUprcNq2bSvff/+9ZGZmSkhIiMP2A/6OIAOgXF1wwQXmjVp7LOzFv8XRAuCUlBRTxKuziPLSdVdCQ0PzHdOAUxwtGNbHFLX6sKvtP3v2rCnYLS6IOXLixAnJyckxPVMF26zBSm96PwDHqJEBUK50po96++23i+wNyTusZA8K+oafl84QynuenfZynD9/vthhIq29yVu/oucuWbLEpfZrbY9+bZ1hVJSi2lXQhRdeaHqTCk4311Cl08e1x8cd69UA/oAeGQDlqlWrVma4SFff1WJarTXR9VROnTolO3fulO+++07Wrl1rztX1Zd566y15/vnnzTRnLZDVYlrt0dHtDgoGob/85S+yePFieeedd0xvj/Z2aAGxBhx9vE6n1uneGqZ0mvOiRYvMSsDOhA87XcBu3bp18uabb8r69etNz4wOfekwmH6uvUS6Dk1JBg0aZOpktD1a+6NT1bWXZsGCBWaNmgceeKCUVxfwPwQZAOVOg4yutTJ//nz54IMPTO9EtWrVTG3M448/nnuehhWtpdHem1mzZpkemjZt2pig8PLLL5sF8PK6//77zfDRxx9/bIKK9uRoj4sGmRtuuEFOnjxp1qV54403TKHwkCFDzHO6MvtJZzFpka6GjuXLl+eGFg1jGtLsCwGWRM/TYl5dS+err74yPUwRERHSunVrszBgwWE0AMVjryUAAGBZ1MgAAADLIsgAAADLIsgAAADLIsgAAADLIsgAAADLIsgAAADLIsgAAADLIsgAAADLIsgAAADLIsgAAADLIsgAAADLIsgAAADLIsgAAACxqv8HOwvlvDk3PlQAAAAASUVORK5CYII=",
      "text/plain": [
       "<Figure size 600x500 with 2 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# Get the index of the feature with the largest mean absolute SHAP value\n",
    "top_feature_idx = np.argmax(np.abs(shap_values_depth1.values).mean(0))\n",
    "shap.plots.scatter(shap_values_depth1[:, top_feature_idx])"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "4dd4641cc4064e0191573fe9c69df29b",
   "metadata": {},
   "source": [
    "**As expected there are no interactions for the depth-1 model:**\n",
    "\n",
    "We can also see this from the lack of any vertical dispersion in the scatter plot above."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "id": "8309879909854d7188b41380fd92a7c3",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2026-04-09T21:21:12.515594Z",
     "iopub.status.busy": "2026-04-09T21:21:12.515455Z",
     "iopub.status.idle": "2026-04-09T21:21:12.640266Z",
     "shell.execute_reply": "2026-04-09T21:21:12.639688Z"
    }
   },
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 750x500 with 3 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# For depth-1 model, interaction values are essentially zero between features\n",
    "# The second feature index (1) for the interaction plot\n",
    "second_feature_idx = 1\n",
    "shap.plots.scatter(shap_values_depth1[:, top_feature_idx], color=shap_values_depth1[:, second_feature_idx])"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "3ed186c9a28b402fb0bc4494df01f08d",
   "metadata": {},
   "source": [
    "**The tail flattening behavior is consistent across all other features.**"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "id": "cb1e1581032b452c9409d6c6813c49d1",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2026-04-09T21:21:12.641970Z",
     "iopub.status.busy": "2026-04-09T21:21:12.641837Z",
     "iopub.status.idle": "2026-04-09T21:21:12.723598Z",
     "shell.execute_reply": "2026-04-09T21:21:12.723051Z"
    }
   },
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 600x500 with 2 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "shap.plots.scatter(shap_values_depth1[:, 1])"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "379cbbc1e968416e875cc15c1202d7eb",
   "metadata": {},
   "source": [
    "**Note that weaker signal lead to more variability in the fits:**\n",
    "\n",
    "Since XGBoost likes flat regions the variability will often look like step functions. Remember in the plot below the SHAP values are correctly telling you what the model learned, but the model did not learn a smooth line."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "id": "277c27b1587741f2af2001be3712ef0d",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2026-04-09T21:21:12.725214Z",
     "iopub.status.busy": "2026-04-09T21:21:12.725087Z",
     "iopub.status.idle": "2026-04-09T21:21:12.805456Z",
     "shell.execute_reply": "2026-04-09T21:21:12.804978Z"
    }
   },
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 600x500 with 2 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "shap.plots.scatter(shap_values_depth1[:, 2])"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "db7b79bc585a40fcaf58bf750017e135",
   "metadata": {},
   "source": [
    "## Explain the depth 3 model\n",
    "\n",
    "The depth 3 model allows for much richer interactions and should do a better job of fitting the linear function."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "id": "916684f9a58a4a2aa5f864670399430d",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2026-04-09T21:21:12.807043Z",
     "iopub.status.busy": "2026-04-09T21:21:12.806899Z",
     "iopub.status.idle": "2026-04-09T21:21:23.895301Z",
     "shell.execute_reply": "2026-04-09T21:21:23.889560Z"
    }
   },
   "outputs": [],
   "source": [
    "explainer_depth3 = shap.TreeExplainer(model_depth3)\n",
    "shap_values_depth3 = explainer_depth3(X_test)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "1671c31a24314836a5b85d7ef7fbf015",
   "metadata": {},
   "source": [
    "### Bar plot for depth 3 model"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "id": "33b0902fd34d4ace834912fa1002cf8e",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2026-04-09T21:21:23.919801Z",
     "iopub.status.busy": "2026-04-09T21:21:23.918705Z",
     "iopub.status.idle": "2026-04-09T21:21:24.171496Z",
     "shell.execute_reply": "2026-04-09T21:21:24.166002Z"
    }
   },
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 800x650 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "shap.plots.bar(shap_values_depth3)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "f6fa52606d8c4a75a9b52967216f8f3f",
   "metadata": {},
   "source": [
    "### Beeswarm plot for depth 3 model"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "id": "f5a1fa73e5044315a093ec459c9be902",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2026-04-09T21:21:24.194855Z",
     "iopub.status.busy": "2026-04-09T21:21:24.193969Z",
     "iopub.status.idle": "2026-04-09T21:21:24.833860Z",
     "shell.execute_reply": "2026-04-09T21:21:24.828268Z"
    }
   },
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 800x550 with 2 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "shap.plots.beeswarm(shap_values_depth3)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "cdf66aed5cc84ca1b48e60bad68798a8",
   "metadata": {},
   "source": [
    "### The depth 3 model fits the linear function much better"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "id": "28d3efd5258a48a79c179ea5c6759f01",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2026-04-09T21:21:24.837957Z",
     "iopub.status.busy": "2026-04-09T21:21:24.837640Z",
     "iopub.status.idle": "2026-04-09T21:21:25.042685Z",
     "shell.execute_reply": "2026-04-09T21:21:25.035345Z"
    }
   },
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 600x500 with 2 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# Get the index of the feature with the largest mean absolute SHAP value for depth 3 model\n",
    "top_feature_idx_d3 = np.argmax(np.abs(shap_values_depth3.values).mean(0))\n",
    "shap.plots.scatter(shap_values_depth3[:, top_feature_idx_d3])"
   ]
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3",
   "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": "3.12.13"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 5
}