{
 "cells": [
  {
   "cell_type": "markdown",
   "id": "c47dm0kxh8d",
   "metadata": {},
   "source": [
    "# Advanced Workflows with nestkit\n",
    "\n",
    "This notebook covers nestkit's advanced analysis tools. It assumes familiarity with the basics covered in **[01  -  Basic Usage](01_basic_usage.ipynb)**.\n",
    "\n",
    "## What you'll learn\n",
    "\n",
    "1. Multi-model statistical comparison with `NestedCVComparator`\n",
    "2. Feature importance aggregation with `FeatureImportanceAggregator`\n",
    "3. Hyperparameter stability diagnostics\n",
    "4. Callbacks: progress, logging, checkpointing\n",
    "\n",
    "## Prerequisites\n",
    "\n",
    "```bash\n",
    "pip install nestkit[full]\n",
    "```"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "id": "sjhslushy89",
   "metadata": {},
   "outputs": [],
   "source": [
    "import logging\n",
    "import os\n",
    "import tempfile\n",
    "import time\n",
    "import warnings\n",
    "\n",
    "import matplotlib.pyplot as plt\n",
    "from sklearn.datasets import load_breast_cancer, load_diabetes\n",
    "from sklearn.ensemble import GradientBoostingClassifier, RandomForestClassifier\n",
    "from sklearn.linear_model import LogisticRegression\n",
    "\n",
    "from nestkit import NestedCVClassifier\n",
    "from nestkit.callbacks import CheckpointCallback, LoggingCallback, ProgressCallback\n",
    "from nestkit.comparison import NestedCVComparator\n",
    "from nestkit.diagnostics import HyperparameterStability\n",
    "from nestkit.importance import FeatureImportanceAggregator\n",
    "from nestkit.plotting import (\n",
    "    plot_bayesian_posterior,\n",
    "    plot_comparison,\n",
    "    plot_critical_difference,\n",
    "    plot_importance,\n",
    "    plot_rank_stability_features,\n",
    "    plot_score_differences,\n",
    "    plot_selection_frequency,\n",
    ")\n",
    "\n",
    "warnings.filterwarnings(\"ignore\")\n",
    "\n",
    "%matplotlib inline\n",
    "plt.rcParams[\"figure.dpi\"] = 100"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "635531ndvm",
   "metadata": {},
   "source": [
    "## 0. Data Setup\n",
    "\n",
    "We reuse the same datasets from notebook 01 to maintain continuity. For multi-model comparison we need all models evaluated on **identical outer folds**, so we fix `outer_cv=5`, `inner_cv=3`, and `random_state=42` throughout."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "id": "amsvasfxou8",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Classification: (569, 30), Regression: (442, 10)\n"
     ]
    }
   ],
   "source": [
    "X_clf, y_clf = load_breast_cancer(return_X_y=True)\n",
    "feature_names_clf = list(load_breast_cancer().feature_names)\n",
    "\n",
    "X_reg, y_reg = load_diabetes(return_X_y=True)\n",
    "feature_names_reg = list(load_diabetes().feature_names)\n",
    "\n",
    "print(f\"Classification: {X_clf.shape}, Regression: {X_reg.shape}\")"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "b7gcqrr3a3w",
   "metadata": {},
   "source": [
    "## 1. Multi-Model Statistical Comparison\n",
    "\n",
    "A common question is: **\"Is model A really better than model B?\"** Simply comparing mean scores is unreliable because outer-fold scores are correlated (overlapping training sets). `NestedCVComparator` provides three principled approaches:\n",
    "\n",
    "1. **Nadeau-Bengio corrected paired t-test**  -  frequentist test that corrects for the non-independence of CV folds\n",
    "2. **Bayesian correlated t-test with ROPE**  -  gives posterior probabilities that model A is better, practically equivalent, or worse\n",
    "3. **Critical difference diagram**  -  visual ranking of 3+ models (Demsar, 2006)\n",
    "\n",
    "### 1.1 Fitting multiple models on identical folds"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "id": "ttqr84lvm5",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "All models fitted.\n"
     ]
    }
   ],
   "source": [
    "# All models share identical outer_cv, inner_cv, random_state\n",
    "common = dict(outer_cv=5, inner_cv=3, scoring=\"roc_auc\", random_state=42)\n",
    "\n",
    "# Model 1: Random Forest\n",
    "ncv_rf = NestedCVClassifier(\n",
    "    estimator=RandomForestClassifier(random_state=42),\n",
    "    param_grid={\"n_estimators\": [50, 100, 200], \"max_depth\": [3, 5, 10]},\n",
    "    **common,\n",
    ")\n",
    "ncv_rf.fit(X_clf, y_clf)\n",
    "\n",
    "# Model 2: Logistic Regression\n",
    "ncv_lr = NestedCVClassifier(\n",
    "    estimator=LogisticRegression(max_iter=5000, random_state=42),\n",
    "    param_grid={\"C\": [0.01, 0.1, 1.0, 10.0], \"penalty\": [\"l1\", \"l2\"], \"solver\": [\"saga\"]},\n",
    "    **common,\n",
    ")\n",
    "ncv_lr.fit(X_clf, y_clf)\n",
    "\n",
    "# Model 3: Gradient Boosting\n",
    "ncv_gb = NestedCVClassifier(\n",
    "    estimator=GradientBoostingClassifier(random_state=42),\n",
    "    param_grid={\"n_estimators\": [50, 100], \"max_depth\": [2, 3, 5], \"learning_rate\": [0.05, 0.1]},\n",
    "    **common,\n",
    ")\n",
    "ncv_gb.fit(X_clf, y_clf)\n",
    "\n",
    "print(\"All models fitted.\")"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "9g24c52spt8",
   "metadata": {},
   "source": [
    "### 1.2 Building the comparator\n",
    "\n",
    "Register each model's results under a human-readable name. `NestedCVComparator` validates that all models share identical outer-fold indices."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "id": "rx101v12ktf",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>model</th>\n",
       "      <th>mean</th>\n",
       "      <th>std</th>\n",
       "      <th>median</th>\n",
       "      <th>ci_lower</th>\n",
       "      <th>ci_upper</th>\n",
       "      <th>min</th>\n",
       "      <th>max</th>\n",
       "      <th>iqr</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>Random Forest</td>\n",
       "      <td>0.990150</td>\n",
       "      <td>0.009537</td>\n",
       "      <td>0.991733</td>\n",
       "      <td>0.972387</td>\n",
       "      <td>1.007913</td>\n",
       "      <td>0.976744</td>\n",
       "      <td>0.999008</td>\n",
       "      <td>0.014053</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>Logistic Regression</td>\n",
       "      <td>0.969716</td>\n",
       "      <td>0.011034</td>\n",
       "      <td>0.970899</td>\n",
       "      <td>0.949166</td>\n",
       "      <td>0.990267</td>\n",
       "      <td>0.956436</td>\n",
       "      <td>0.985260</td>\n",
       "      <td>0.011104</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>Gradient Boosting</td>\n",
       "      <td>0.991837</td>\n",
       "      <td>0.006119</td>\n",
       "      <td>0.990829</td>\n",
       "      <td>0.980441</td>\n",
       "      <td>1.003233</td>\n",
       "      <td>0.985450</td>\n",
       "      <td>0.998347</td>\n",
       "      <td>0.011417</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                 model      mean       std    median  ci_lower  ci_upper  \\\n",
       "0        Random Forest  0.990150  0.009537  0.991733  0.972387  1.007913   \n",
       "1  Logistic Regression  0.969716  0.011034  0.970899  0.949166  0.990267   \n",
       "2    Gradient Boosting  0.991837  0.006119  0.990829  0.980441  1.003233   \n",
       "\n",
       "        min       max       iqr  \n",
       "0  0.976744  0.999008  0.014053  \n",
       "1  0.956436  0.985260  0.011104  \n",
       "2  0.985450  0.998347  0.011417  "
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "comparator = NestedCVComparator()\n",
    "comparator.add(\"Random Forest\", ncv_rf.results_)\n",
    "comparator.add(\"Logistic Regression\", ncv_lr.results_)\n",
    "comparator.add(\"Gradient Boosting\", ncv_gb.results_)\n",
    "\n",
    "display(comparator.summary(\"roc_auc\"))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "id": "cdrp6hbhvc",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>model</th>\n",
       "      <th>mean</th>\n",
       "      <th>std</th>\n",
       "      <th>median</th>\n",
       "      <th>ci_lower</th>\n",
       "      <th>ci_upper</th>\n",
       "      <th>min</th>\n",
       "      <th>max</th>\n",
       "      <th>iqr</th>\n",
       "      <th>rank</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>Gradient Boosting</td>\n",
       "      <td>0.991837</td>\n",
       "      <td>0.006119</td>\n",
       "      <td>0.990829</td>\n",
       "      <td>0.980441</td>\n",
       "      <td>1.003233</td>\n",
       "      <td>0.985450</td>\n",
       "      <td>0.998347</td>\n",
       "      <td>0.011417</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>Random Forest</td>\n",
       "      <td>0.990150</td>\n",
       "      <td>0.009537</td>\n",
       "      <td>0.991733</td>\n",
       "      <td>0.972387</td>\n",
       "      <td>1.007913</td>\n",
       "      <td>0.976744</td>\n",
       "      <td>0.999008</td>\n",
       "      <td>0.014053</td>\n",
       "      <td>2</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>Logistic Regression</td>\n",
       "      <td>0.969716</td>\n",
       "      <td>0.011034</td>\n",
       "      <td>0.970899</td>\n",
       "      <td>0.949166</td>\n",
       "      <td>0.990267</td>\n",
       "      <td>0.956436</td>\n",
       "      <td>0.985260</td>\n",
       "      <td>0.011104</td>\n",
       "      <td>3</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                 model      mean       std    median  ci_lower  ci_upper  \\\n",
       "0    Gradient Boosting  0.991837  0.006119  0.990829  0.980441  1.003233   \n",
       "1        Random Forest  0.990150  0.009537  0.991733  0.972387  1.007913   \n",
       "2  Logistic Regression  0.969716  0.011034  0.970899  0.949166  0.990267   \n",
       "\n",
       "        min       max       iqr  rank  \n",
       "0  0.985450  0.998347  0.011417     1  \n",
       "1  0.976744  0.999008  0.014053     2  \n",
       "2  0.956436  0.985260  0.011104     3  "
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# Rank models by mean ROC AUC\n",
    "display(comparator.rank_models(\"roc_auc\"))"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "7oewjh7pbmx",
   "metadata": {},
   "source": [
    "### 1.3 Visual comparison\n",
    "\n",
    "`plot_comparison` draws paired box-and-strip plots with lines connecting the same fold across models  -  making it easy to spot whether one model consistently wins."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "id": "nj3wn4lf1k",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "plot_comparison(comparator, \"roc_auc\", line_alpha=0.1)\n",
    "plt.tight_layout()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "81q27v4z6l",
   "metadata": {},
   "source": [
    "### 1.4 Nadeau-Bengio corrected paired t-test\n",
    "\n",
    "Standard paired t-tests underestimate the variance of CV scores because training sets overlap. The Nadeau-Bengio correction (2003) adjusts the variance estimate by accounting for the ratio of test-set size to training-set size."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "id": "hlt1ewckduv",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "RF vs LR (Nadeau-Bengio corrected):\n",
      "  t_statistic: 2.212985001740397\n",
      "  p_value: 0.0913221564679729\n",
      "  mean_difference: 0.020433280128661656\n",
      "  corrected_std: 0.009233356806572095\n",
      "  ci_lower: -0.005202628181490216\n",
      "  ci_upper: 0.046069188438813524\n",
      "  n_folds: 5\n",
      "  significant_at_005: False\n",
      "  significant_at_001: False\n"
     ]
    }
   ],
   "source": [
    "# Single pairwise test\n",
    "result = comparator.corrected_paired_ttest(\"roc_auc\", \"Random Forest\", \"Logistic Regression\")\n",
    "print(\"RF vs LR (Nadeau-Bengio corrected):\")\n",
    "for k, v in result.items():\n",
    "    print(f\"  {k}: {v}\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "id": "bhgwxor272d",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>model_a</th>\n",
       "      <th>model_b</th>\n",
       "      <th>t_statistic</th>\n",
       "      <th>p_value</th>\n",
       "      <th>mean_difference</th>\n",
       "      <th>corrected_std</th>\n",
       "      <th>ci_lower</th>\n",
       "      <th>ci_upper</th>\n",
       "      <th>n_folds</th>\n",
       "      <th>significant_at_005</th>\n",
       "      <th>significant_at_001</th>\n",
       "      <th>p_value_corrected</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>Random Forest</td>\n",
       "      <td>Logistic Regression</td>\n",
       "      <td>2.212985</td>\n",
       "      <td>0.091322</td>\n",
       "      <td>0.020433</td>\n",
       "      <td>0.009233</td>\n",
       "      <td>-0.005203</td>\n",
       "      <td>0.046069</td>\n",
       "      <td>5</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>0.273966</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>Random Forest</td>\n",
       "      <td>Gradient Boosting</td>\n",
       "      <td>-0.332813</td>\n",
       "      <td>0.755988</td>\n",
       "      <td>-0.001687</td>\n",
       "      <td>0.005069</td>\n",
       "      <td>-0.015760</td>\n",
       "      <td>0.012386</td>\n",
       "      <td>5</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>0.755988</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>Logistic Regression</td>\n",
       "      <td>Gradient Boosting</td>\n",
       "      <td>-2.202411</td>\n",
       "      <td>0.092404</td>\n",
       "      <td>-0.022120</td>\n",
       "      <td>0.010044</td>\n",
       "      <td>-0.050006</td>\n",
       "      <td>0.005765</td>\n",
       "      <td>5</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>0.273966</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "               model_a              model_b  t_statistic   p_value  \\\n",
       "0        Random Forest  Logistic Regression     2.212985  0.091322   \n",
       "1        Random Forest    Gradient Boosting    -0.332813  0.755988   \n",
       "2  Logistic Regression    Gradient Boosting    -2.202411  0.092404   \n",
       "\n",
       "   mean_difference  corrected_std  ci_lower  ci_upper  n_folds  \\\n",
       "0         0.020433       0.009233 -0.005203  0.046069        5   \n",
       "1        -0.001687       0.005069 -0.015760  0.012386        5   \n",
       "2        -0.022120       0.010044 -0.050006  0.005765        5   \n",
       "\n",
       "   significant_at_005  significant_at_001  p_value_corrected  \n",
       "0               False               False           0.273966  \n",
       "1               False               False           0.755988  \n",
       "2               False               False           0.273966  "
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# All pairwise tests with Holm-Bonferroni correction for multiple comparisons\n",
    "display(comparator.pairwise_corrected_ttest(\"roc_auc\"))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "id": "wyhc8knlkb",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "plot_score_differences(comparator, \"roc_auc\", \"Random Forest\", \"Logistic Regression\")\n",
    "plt.tight_layout()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "6noi331ld6",
   "metadata": {},
   "source": [
    "### 1.5 Bayesian correlated t-test with ROPE\n",
    "\n",
    "The Bayesian test (Benavoli et al., 2017) partitions the posterior probability mass into three regions:\n",
    "- **P(A better)**  -  model A's mean score exceeds model B's by more than the ROPE\n",
    "- **P(equivalent)**  -  the difference lies within the Region of Practical Equivalence\n",
    "- **P(B better)**  -  model B's mean score exceeds model A's by more than the ROPE\n",
    "\n",
    "This avoids the binary \"significant / not significant\" dichotomy and quantifies practical equivalence."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "id": "rq6f94dceo",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Bayesian comparison (RF vs GB):\n",
      "  p_a_better: 0.0412\n",
      "  p_equivalent: 0.8706\n",
      "  p_b_better: 0.0882\n",
      "  rope: 0.0100\n",
      "  mean_difference: -0.0017\n",
      "  hdi_lower: -0.0158\n",
      "  hdi_upper: 0.0124\n"
     ]
    }
   ],
   "source": [
    "bayes = comparator.bayesian_comparison(\"roc_auc\", \"Random Forest\", \"Gradient Boosting\", rope=0.01)\n",
    "print(\"Bayesian comparison (RF vs GB):\")\n",
    "for k, v in bayes.items():\n",
    "    print(f\"  {k}: {v:.4f}\" if isinstance(v, float) else f\"  {k}: {v}\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "id": "kt449d7kye8",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "plot_bayesian_posterior(comparator, \"roc_auc\", \"Random Forest\", \"Gradient Boosting\", rope=0.01)\n",
    "plt.tight_layout()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "6ju535g3kac",
   "metadata": {},
   "source": [
    "### 1.6 Critical difference diagram\n",
    "\n",
    "When comparing three or more models, a critical difference diagram (Demsar, 2006) ranks models by their average rank across folds and groups models that are not significantly different."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "id": "ke3co4t7dle",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "plot_critical_difference(comparator, \"roc_auc\")\n",
    "plt.tight_layout()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "1m3fc5tgy0p",
   "metadata": {},
   "source": [
    "## 2. Feature Importance Analysis\n",
    "\n",
    "Feature importances from a single train/test split can be noisy and misleading. `FeatureImportanceAggregator` extracts importances from each outer-fold estimator, aggregates them, and assesses their **stability**  -  how consistently the same features are ranked important across folds.\n",
    "\n",
    "Key outputs:\n",
    "- **Summary table**  -  mean, std, CV, and rank statistics per feature\n",
    "- **Nogueira stability index**  -  measures agreement of top-k feature sets across folds (1 = perfect, 0 = random)\n",
    "- **Consensus features**  -  features that reliably appear in the top-k across folds"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "id": "hnxr8jp3t4j",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Top 10 features by mean importance:\n"
     ]
    },
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>feature</th>\n",
       "      <th>mean_importance</th>\n",
       "      <th>std_importance</th>\n",
       "      <th>median_importance</th>\n",
       "      <th>min_importance</th>\n",
       "      <th>max_importance</th>\n",
       "      <th>cv_importance</th>\n",
       "      <th>mean_rank</th>\n",
       "      <th>std_rank</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>worst area</td>\n",
       "      <td>0.146445</td>\n",
       "      <td>0.032512</td>\n",
       "      <td>0.154264</td>\n",
       "      <td>0.092618</td>\n",
       "      <td>0.180883</td>\n",
       "      <td>0.222011</td>\n",
       "      <td>1.6</td>\n",
       "      <td>1.341641</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>worst concave points</td>\n",
       "      <td>0.128764</td>\n",
       "      <td>0.008015</td>\n",
       "      <td>0.126759</td>\n",
       "      <td>0.120410</td>\n",
       "      <td>0.141930</td>\n",
       "      <td>0.062244</td>\n",
       "      <td>2.2</td>\n",
       "      <td>0.836660</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>worst perimeter</td>\n",
       "      <td>0.104169</td>\n",
       "      <td>0.030896</td>\n",
       "      <td>0.094333</td>\n",
       "      <td>0.076230</td>\n",
       "      <td>0.148497</td>\n",
       "      <td>0.296590</td>\n",
       "      <td>3.6</td>\n",
       "      <td>1.816590</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>mean concave points</td>\n",
       "      <td>0.095041</td>\n",
       "      <td>0.012074</td>\n",
       "      <td>0.093302</td>\n",
       "      <td>0.080057</td>\n",
       "      <td>0.113452</td>\n",
       "      <td>0.127036</td>\n",
       "      <td>3.4</td>\n",
       "      <td>0.894427</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>worst radius</td>\n",
       "      <td>0.074889</td>\n",
       "      <td>0.008408</td>\n",
       "      <td>0.079846</td>\n",
       "      <td>0.061508</td>\n",
       "      <td>0.080985</td>\n",
       "      <td>0.112277</td>\n",
       "      <td>4.8</td>\n",
       "      <td>1.303840</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5</th>\n",
       "      <td>mean concavity</td>\n",
       "      <td>0.066027</td>\n",
       "      <td>0.008330</td>\n",
       "      <td>0.063741</td>\n",
       "      <td>0.057992</td>\n",
       "      <td>0.079635</td>\n",
       "      <td>0.126155</td>\n",
       "      <td>6.4</td>\n",
       "      <td>0.894427</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>6</th>\n",
       "      <td>mean perimeter</td>\n",
       "      <td>0.062685</td>\n",
       "      <td>0.016754</td>\n",
       "      <td>0.070298</td>\n",
       "      <td>0.043445</td>\n",
       "      <td>0.078647</td>\n",
       "      <td>0.267271</td>\n",
       "      <td>7.0</td>\n",
       "      <td>1.581139</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>7</th>\n",
       "      <td>mean area</td>\n",
       "      <td>0.048882</td>\n",
       "      <td>0.012535</td>\n",
       "      <td>0.048680</td>\n",
       "      <td>0.030149</td>\n",
       "      <td>0.064769</td>\n",
       "      <td>0.256436</td>\n",
       "      <td>8.8</td>\n",
       "      <td>1.303840</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>8</th>\n",
       "      <td>mean radius</td>\n",
       "      <td>0.041921</td>\n",
       "      <td>0.006607</td>\n",
       "      <td>0.043092</td>\n",
       "      <td>0.032296</td>\n",
       "      <td>0.049127</td>\n",
       "      <td>0.157613</td>\n",
       "      <td>9.4</td>\n",
       "      <td>1.816590</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>9</th>\n",
       "      <td>worst concavity</td>\n",
       "      <td>0.040636</td>\n",
       "      <td>0.013206</td>\n",
       "      <td>0.034822</td>\n",
       "      <td>0.029775</td>\n",
       "      <td>0.062906</td>\n",
       "      <td>0.324974</td>\n",
       "      <td>9.4</td>\n",
       "      <td>2.509980</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                feature  mean_importance  std_importance  median_importance  \\\n",
       "0            worst area         0.146445        0.032512           0.154264   \n",
       "1  worst concave points         0.128764        0.008015           0.126759   \n",
       "2       worst perimeter         0.104169        0.030896           0.094333   \n",
       "3   mean concave points         0.095041        0.012074           0.093302   \n",
       "4          worst radius         0.074889        0.008408           0.079846   \n",
       "5        mean concavity         0.066027        0.008330           0.063741   \n",
       "6        mean perimeter         0.062685        0.016754           0.070298   \n",
       "7             mean area         0.048882        0.012535           0.048680   \n",
       "8           mean radius         0.041921        0.006607           0.043092   \n",
       "9       worst concavity         0.040636        0.013206           0.034822   \n",
       "\n",
       "   min_importance  max_importance  cv_importance  mean_rank  std_rank  \n",
       "0        0.092618        0.180883       0.222011        1.6  1.341641  \n",
       "1        0.120410        0.141930       0.062244        2.2  0.836660  \n",
       "2        0.076230        0.148497       0.296590        3.6  1.816590  \n",
       "3        0.080057        0.113452       0.127036        3.4  0.894427  \n",
       "4        0.061508        0.080985       0.112277        4.8  1.303840  \n",
       "5        0.057992        0.079635       0.126155        6.4  0.894427  \n",
       "6        0.043445        0.078647       0.267271        7.0  1.581139  \n",
       "7        0.030149        0.064769       0.256436        8.8  1.303840  \n",
       "8        0.032296        0.049127       0.157613        9.4  1.816590  \n",
       "9        0.029775        0.062906       0.324974        9.4  2.509980  "
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "agg = FeatureImportanceAggregator(\n",
    "    ncv_rf.results_,\n",
    "    method=\"model\",\n",
    "    feature_names=feature_names_clf,\n",
    ")\n",
    "agg.compute()\n",
    "\n",
    "print(\"Top 10 features by mean importance:\")\n",
    "display(agg.summary_.head(10))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "id": "t9jgs093w8s",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 1400x600 with 2 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "fig, axes = plt.subplots(1, 2, figsize=(14, 6))\n",
    "plot_importance(agg, top_k=15, ax=axes[0])\n",
    "plot_selection_frequency(agg, top_k=10, ax=axes[1])\n",
    "plt.tight_layout()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "orwcoash42",
   "metadata": {},
   "source": [
    "### 2.1 Nogueira stability index\n",
    "\n",
    "The Nogueira et al. (2018) stability index measures how consistently the same features appear in the top-k set across folds, corrected for chance agreement. Values near 1 indicate high stability; values near 0 indicate that top features vary across folds."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "id": "qd2dntp3ek",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Stability index (top-5): 0.7360\n",
      "Stability index (top-10): 0.8650\n",
      "Stability index (top-15): 0.7867\n"
     ]
    }
   ],
   "source": [
    "for k in [5, 10, 15]:\n",
    "    idx = agg.stability_index(top_k=k)\n",
    "    print(f\"Stability index (top-{k}): {idx:.4f}\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "id": "eygxk37lxvf",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 640x480 with 2 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "plot_rank_stability_features(agg, top_k=15)\n",
    "plt.tight_layout()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "6xgqz1z5pnj",
   "metadata": {},
   "source": [
    "### 2.2 Consensus features\n",
    "\n",
    "Two strategies for identifying reliably important features:\n",
    "- `\"top_k\"`  -  the top-k features by mean importance\n",
    "- `\"frequency\"`  -  features appearing in the top-k in at least a given fraction of folds (more conservative)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "id": "6cc3tegsouu",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Top-10 by mean importance:\n",
      "  [np.str_('worst area'), np.str_('worst concave points'), np.str_('worst perimeter'), np.str_('mean concave points'), np.str_('worst radius'), np.str_('mean concavity'), np.str_('mean perimeter'), np.str_('mean area'), np.str_('mean radius'), np.str_('worst concavity')]\n",
      "\n",
      "Consensus features (top-10, >=80% folds):\n",
      "  [np.str_('mean radius'), np.str_('mean perimeter'), np.str_('mean area'), np.str_('mean concavity'), np.str_('mean concave points'), np.str_('area error'), np.str_('worst radius'), np.str_('worst perimeter'), np.str_('worst area'), np.str_('worst concave points')]\n"
     ]
    }
   ],
   "source": [
    "# Top-10 by mean importance\n",
    "top10 = agg.consensus_features(criterion=\"top_k\", top_k=10)\n",
    "print(f\"Top-10 by mean importance:\\n  {top10}\\n\")\n",
    "\n",
    "# Features in top-10 in at least 80% of folds\n",
    "consensus = agg.consensus_features(criterion=\"frequency\", top_k=10, min_frequency=0.8)\n",
    "print(f\"Consensus features (top-10, >=80% folds):\\n  {consensus}\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "id": "k5pvrh9y6ce",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Pairwise Spearman rank correlation:\n"
     ]
    },
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>fold_i</th>\n",
       "      <th>fold_j</th>\n",
       "      <th>spearman_r</th>\n",
       "      <th>p_value</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0.933704</td>\n",
       "      <td>5.157534e-14</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>0</td>\n",
       "      <td>2</td>\n",
       "      <td>0.920801</td>\n",
       "      <td>5.735790e-13</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>0</td>\n",
       "      <td>3</td>\n",
       "      <td>0.954171</td>\n",
       "      <td>3.333087e-16</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>0</td>\n",
       "      <td>4</td>\n",
       "      <td>0.926140</td>\n",
       "      <td>2.232463e-13</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>1</td>\n",
       "      <td>2</td>\n",
       "      <td>0.939488</td>\n",
       "      <td>1.489555e-14</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5</th>\n",
       "      <td>1</td>\n",
       "      <td>3</td>\n",
       "      <td>0.948832</td>\n",
       "      <td>1.508414e-15</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>6</th>\n",
       "      <td>1</td>\n",
       "      <td>4</td>\n",
       "      <td>0.964850</td>\n",
       "      <td>8.680629e-18</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>7</th>\n",
       "      <td>2</td>\n",
       "      <td>3</td>\n",
       "      <td>0.948387</td>\n",
       "      <td>1.698100e-15</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>8</th>\n",
       "      <td>2</td>\n",
       "      <td>4</td>\n",
       "      <td>0.953726</td>\n",
       "      <td>3.805360e-16</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>9</th>\n",
       "      <td>3</td>\n",
       "      <td>4</td>\n",
       "      <td>0.952836</td>\n",
       "      <td>4.941027e-16</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   fold_i  fold_j  spearman_r       p_value\n",
       "0       0       1    0.933704  5.157534e-14\n",
       "1       0       2    0.920801  5.735790e-13\n",
       "2       0       3    0.954171  3.333087e-16\n",
       "3       0       4    0.926140  2.232463e-13\n",
       "4       1       2    0.939488  1.489555e-14\n",
       "5       1       3    0.948832  1.508414e-15\n",
       "6       1       4    0.964850  8.680629e-18\n",
       "7       2       3    0.948387  1.698100e-15\n",
       "8       2       4    0.953726  3.805360e-16\n",
       "9       3       4    0.952836  4.941027e-16"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# Pairwise Spearman rank correlation between folds\n",
    "print(\"Pairwise Spearman rank correlation:\")\n",
    "display(agg.pairwise_rank_correlation())"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "ae2puei2lj6",
   "metadata": {},
   "source": [
    "## 3. Hyperparameter Stability Diagnostics\n",
    "\n",
    "If the inner CV selects different hyperparameters in different outer folds, the model may be sensitive to the training data composition. `HyperparameterStability` diagnoses this by computing:\n",
    "\n",
    "- **Mode and agreement rate**  -  how often each parameter's most-common value was selected\n",
    "- **Entropy**  -  low entropy means consistent selection\n",
    "- **Pairwise Jaccard similarity**  -  how similar the full configuration sets are across folds"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "id": "72c43x8vbps",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "RF hyperparameter stability summary:\n"
     ]
    },
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>param</th>\n",
       "      <th>mode</th>\n",
       "      <th>nunique</th>\n",
       "      <th>entropy</th>\n",
       "      <th>agreement_rate</th>\n",
       "      <th>cv</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>max_depth</td>\n",
       "      <td>5</td>\n",
       "      <td>2</td>\n",
       "      <td>0.721928</td>\n",
       "      <td>0.8</td>\n",
       "      <td>0.372678</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>n_estimators</td>\n",
       "      <td>100</td>\n",
       "      <td>3</td>\n",
       "      <td>1.521928</td>\n",
       "      <td>0.4</td>\n",
       "      <td>0.516016</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "          param mode  nunique   entropy  agreement_rate        cv\n",
       "0     max_depth    5        2  0.721928             0.8  0.372678\n",
       "1  n_estimators  100        3  1.521928             0.4  0.516016"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "hs_rf = HyperparameterStability(ncv_rf.results_.best_params_per_fold_)\n",
    "\n",
    "print(\"RF hyperparameter stability summary:\")\n",
    "display(hs_rf.summary())"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "id": "qazrui1xo7",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Stable parameters: {'max_depth': True, 'n_estimators': False}\n",
      "\n",
      "Pairwise Jaccard similarity between fold configurations:\n"
     ]
    },
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>fold_i</th>\n",
       "      <th>fold_j</th>\n",
       "      <th>jaccard</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>0</td>\n",
       "      <td>2</td>\n",
       "      <td>0.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>0</td>\n",
       "      <td>3</td>\n",
       "      <td>0.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>0</td>\n",
       "      <td>4</td>\n",
       "      <td>0.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>1</td>\n",
       "      <td>2</td>\n",
       "      <td>0.333333</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5</th>\n",
       "      <td>1</td>\n",
       "      <td>3</td>\n",
       "      <td>0.333333</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>6</th>\n",
       "      <td>1</td>\n",
       "      <td>4</td>\n",
       "      <td>1.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>7</th>\n",
       "      <td>2</td>\n",
       "      <td>3</td>\n",
       "      <td>1.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>8</th>\n",
       "      <td>2</td>\n",
       "      <td>4</td>\n",
       "      <td>0.333333</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>9</th>\n",
       "      <td>3</td>\n",
       "      <td>4</td>\n",
       "      <td>0.333333</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   fold_i  fold_j   jaccard\n",
       "0       0       1  0.000000\n",
       "1       0       2  0.000000\n",
       "2       0       3  0.000000\n",
       "3       0       4  0.000000\n",
       "4       1       2  0.333333\n",
       "5       1       3  0.333333\n",
       "6       1       4  1.000000\n",
       "7       2       3  1.000000\n",
       "8       2       4  0.333333\n",
       "9       3       4  0.333333"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# Which parameters are stable (>=80% agreement)?\n",
    "print(\"Stable parameters:\", hs_rf.is_stable(threshold=0.8))\n",
    "\n",
    "print(\"\\nPairwise Jaccard similarity between fold configurations:\")\n",
    "display(hs_rf.pairwise_jaccard())"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "id": "g8fm6s42o8p",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "RF: mean agreement=0.60, mean entropy=1.12\n",
      "LR: mean agreement=0.80, mean entropy=0.56\n",
      "GB: mean agreement=0.80, mean entropy=0.56\n"
     ]
    }
   ],
   "source": [
    "# Compare stability across all three models\n",
    "for name, ncv in [(\"RF\", ncv_rf), (\"LR\", ncv_lr), (\"GB\", ncv_gb)]:\n",
    "    hs = HyperparameterStability(ncv.results_.best_params_per_fold_)\n",
    "    summary = hs.summary()\n",
    "    mean_agreement = summary[\"agreement_rate\"].mean()\n",
    "    mean_entropy = summary[\"entropy\"].mean()\n",
    "    print(f\"{name}: mean agreement={mean_agreement:.2f}, mean entropy={mean_entropy:.2f}\")"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "5ae9fc23",
   "metadata": {},
   "source": [
    "## 4. Conformal Prediction\n",
    "\n",
    "nestkit supports **CV+ Mondrian conformal prediction** for both classification and regression.\n",
    "\n",
    "- **Classifier**: per-class conformal quantile thresholds produce *prediction sets* with formal coverage guarantees\n",
    "- **Regressor**: Mondrian binning by predicted value yields *conditional prediction intervals* that are tighter in easy-to-predict regions\n",
    "\n",
    "Conformal prediction runs after calibration and threshold optimization, using calibrated OOF probabilities when available."
   ]
  },
  {
   "cell_type": "markdown",
   "id": "1f88a8d9",
   "metadata": {},
   "source": [
    "### 4.1 Classifier conformal prediction sets\n",
    "\n",
    "Enable conformal prediction with `conformal_prediction=True`. The `conformal_alpha` parameter controls the target miscoverage rate (default 0.1 = 90% target coverage)."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "id": "8c019c14",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Conformal prediction enabled: True\n",
      "Mean coverage: 0.907\n",
      "Mean set size: 0.924\n"
     ]
    }
   ],
   "source": [
    "ncv_conformal = NestedCVClassifier(\n",
    "    estimator=RandomForestClassifier(random_state=42),\n",
    "    param_grid={\"n_estimators\": [50, 100], \"max_depth\": [3, 5]},\n",
    "    outer_cv=5,\n",
    "    inner_cv=3,\n",
    "    scoring=\"roc_auc\",\n",
    "    random_state=42,\n",
    "    conformal_prediction=True,\n",
    "    conformal_alpha=0.1,\n",
    ")\n",
    "ncv_conformal.fit(X_clf, y_clf)\n",
    "\n",
    "print(\"Conformal prediction enabled:\", ncv_conformal.results_.has_conformal)\n",
    "print(f\"Mean coverage: {ncv_conformal.results_.conformal_coverage_['mean']:.3f}\")\n",
    "print(f\"Mean set size: {ncv_conformal.results_.conformal_set_size_stats_['mean']:.3f}\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "id": "822a666f",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>fold_idx</th>\n",
       "      <th>coverage</th>\n",
       "      <th>mean_set_size</th>\n",
       "      <th>frac_singleton</th>\n",
       "      <th>frac_empty</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>0</td>\n",
       "      <td>0.877193</td>\n",
       "      <td>0.912281</td>\n",
       "      <td>0.912281</td>\n",
       "      <td>0.087719</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>1</td>\n",
       "      <td>0.868421</td>\n",
       "      <td>0.894737</td>\n",
       "      <td>0.894737</td>\n",
       "      <td>0.105263</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>2</td>\n",
       "      <td>0.964912</td>\n",
       "      <td>0.973684</td>\n",
       "      <td>0.973684</td>\n",
       "      <td>0.026316</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>3</td>\n",
       "      <td>0.894737</td>\n",
       "      <td>0.903509</td>\n",
       "      <td>0.903509</td>\n",
       "      <td>0.096491</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>4</td>\n",
       "      <td>0.929204</td>\n",
       "      <td>0.938053</td>\n",
       "      <td>0.938053</td>\n",
       "      <td>0.061947</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   fold_idx  coverage  mean_set_size  frac_singleton  frac_empty\n",
       "0         0  0.877193       0.912281        0.912281    0.087719\n",
       "1         1  0.868421       0.894737        0.894737    0.105263\n",
       "2         2  0.964912       0.973684        0.973684    0.026316\n",
       "3         3  0.894737       0.903509        0.903509    0.096491\n",
       "4         4  0.929204       0.938053        0.938053    0.061947"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# Per-fold conformal report\n",
    "display(ncv_conformal.results_.conformal_report())"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 24,
   "id": "94b072f9",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>y_true</th>\n",
       "      <th>y_pred_default</th>\n",
       "      <th>conformal_set_size</th>\n",
       "      <th>conformal_in_set</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>True</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>True</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>True</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>True</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>True</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5</th>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>True</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>6</th>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>True</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>7</th>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>True</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>8</th>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>True</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>9</th>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>True</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   y_true  y_pred_default  conformal_set_size  conformal_in_set\n",
       "0       0               0                   1              True\n",
       "1       0               0                   1              True\n",
       "2       0               0                   1              True\n",
       "3       0               0                   1              True\n",
       "4       0               0                   1              True\n",
       "5       0               0                   1              True\n",
       "6       0               0                   1              True\n",
       "7       0               0                   1              True\n",
       "8       0               0                   1              True\n",
       "9       0               0                   1              True"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# Predictions DataFrame now includes conformal_set_size and conformal_in_set\n",
    "display(\n",
    "    ncv_conformal.results_.predictions_[\n",
    "        [\"y_true\", \"y_pred_default\", \"conformal_set_size\", \"conformal_in_set\"]\n",
    "    ].head(10)\n",
    ")"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "4a558f3f",
   "metadata": {},
   "source": [
    "### 4.2 Combining calibration + threshold + conformal\n",
    "\n",
    "All three post-hoc features compose naturally. When calibration is enabled, conformal thresholds are computed from calibrated probabilities."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 25,
   "id": "4da4ac00",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Calibration: True, Threshold: True, Conformal: True\n",
      "Mean conformal coverage: 0.928\n",
      "q-hat stability (std per class): [0.14066273709105082, 0.07630643735352548]\n"
     ]
    }
   ],
   "source": [
    "ncv_full = NestedCVClassifier(\n",
    "    estimator=RandomForestClassifier(random_state=42),\n",
    "    param_grid={\"n_estimators\": [50, 100], \"max_depth\": [3, 5]},\n",
    "    outer_cv=5,\n",
    "    inner_cv=3,\n",
    "    scoring=\"roc_auc\",\n",
    "    random_state=42,\n",
    "    calibration_method=\"isotonic\",\n",
    "    threshold_strategy=\"pooled\",\n",
    "    conformal_prediction=True,\n",
    "    conformal_alpha=0.1,\n",
    ")\n",
    "ncv_full.fit(X_clf, y_clf)\n",
    "\n",
    "r = ncv_full.results_\n",
    "print(\n",
    "    f\"Calibration: {r.has_calibration}, Threshold: {r.has_threshold_optimization}, Conformal: {r.has_conformal}\"\n",
    ")\n",
    "print(f\"Mean conformal coverage: {r.conformal_coverage_['mean']:.3f}\")\n",
    "print(f\"q-hat stability (std per class): {r.conformal_qhat_stability_['std_per_class']}\")"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "84f0a830",
   "metadata": {},
   "source": [
    "### 4.3 Regressor: Mondrian prediction intervals\n",
    "\n",
    "For regression, `mondrian_bins` enables Mondrian-style conditional prediction intervals. OOF predictions are grouped into equal-frequency bins, and per-bin residual quantiles give tighter intervals in easy-to-predict regions."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 26,
   "id": "ffb7ef84",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Overall coverage: 0.909\n",
      "Per-bin coverage: {0: 0.9054054054054054, 1: 0.9266666666666666, 2: 0.8958333333333334}\n"
     ]
    }
   ],
   "source": [
    "from sklearn.ensemble import RandomForestRegressor\n",
    "\n",
    "from nestkit import NestedCVRegressor\n",
    "\n",
    "ncv_reg_mondrian = NestedCVRegressor(\n",
    "    estimator=RandomForestRegressor(random_state=42),\n",
    "    param_grid={\"n_estimators\": [50, 100], \"max_depth\": [5, 10]},\n",
    "    outer_cv=5,\n",
    "    inner_cv=3,\n",
    "    random_state=42,\n",
    "    prediction_intervals=True,\n",
    "    confidence_level=0.90,\n",
    "    mondrian_bins=3,\n",
    ")\n",
    "ncv_reg_mondrian.fit(X_reg, y_reg)\n",
    "\n",
    "r_reg = ncv_reg_mondrian.results_\n",
    "print(f\"Overall coverage: {r_reg.prediction_interval_coverage_['mean']:.3f}\")\n",
    "print(f\"Per-bin coverage: {r_reg.mondrian_coverage_per_bin_}\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 27,
   "id": "48ef2edb",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>y_true</th>\n",
       "      <th>y_pred</th>\n",
       "      <th>pi_lower</th>\n",
       "      <th>pi_upper</th>\n",
       "      <th>mondrian_bin</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>151.0</td>\n",
       "      <td>215.932084</td>\n",
       "      <td>106.851684</td>\n",
       "      <td>303.158261</td>\n",
       "      <td>2</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>75.0</td>\n",
       "      <td>85.428902</td>\n",
       "      <td>29.613270</td>\n",
       "      <td>188.323636</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>141.0</td>\n",
       "      <td>181.896152</td>\n",
       "      <td>72.815752</td>\n",
       "      <td>269.122328</td>\n",
       "      <td>2</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>206.0</td>\n",
       "      <td>157.863997</td>\n",
       "      <td>72.079456</td>\n",
       "      <td>294.932256</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>135.0</td>\n",
       "      <td>95.199586</td>\n",
       "      <td>39.383955</td>\n",
       "      <td>198.094321</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5</th>\n",
       "      <td>97.0</td>\n",
       "      <td>102.992719</td>\n",
       "      <td>47.177088</td>\n",
       "      <td>205.887454</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>6</th>\n",
       "      <td>138.0</td>\n",
       "      <td>82.646132</td>\n",
       "      <td>26.830501</td>\n",
       "      <td>185.540867</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>7</th>\n",
       "      <td>63.0</td>\n",
       "      <td>142.896578</td>\n",
       "      <td>57.112037</td>\n",
       "      <td>279.964837</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>8</th>\n",
       "      <td>110.0</td>\n",
       "      <td>157.771662</td>\n",
       "      <td>71.987120</td>\n",
       "      <td>294.839921</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>9</th>\n",
       "      <td>310.0</td>\n",
       "      <td>174.042726</td>\n",
       "      <td>88.258184</td>\n",
       "      <td>311.110985</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   y_true      y_pred    pi_lower    pi_upper  mondrian_bin\n",
       "0   151.0  215.932084  106.851684  303.158261             2\n",
       "1    75.0   85.428902   29.613270  188.323636             0\n",
       "2   141.0  181.896152   72.815752  269.122328             2\n",
       "3   206.0  157.863997   72.079456  294.932256             1\n",
       "4   135.0   95.199586   39.383955  198.094321             0\n",
       "5    97.0  102.992719   47.177088  205.887454             0\n",
       "6   138.0   82.646132   26.830501  185.540867             0\n",
       "7    63.0  142.896578   57.112037  279.964837             1\n",
       "8   110.0  157.771662   71.987120  294.839921             1\n",
       "9   310.0  174.042726   88.258184  311.110985             1"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# Predictions include per-sample bin assignments and intervals\n",
    "display(\n",
    "    ncv_reg_mondrian.results_.predictions_[\n",
    "        [\"y_true\", \"y_pred\", \"pi_lower\", \"pi_upper\", \"mondrian_bin\"]\n",
    "    ].head(10)\n",
    ")"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "56wg6yys7to",
   "metadata": {},
   "source": [
    "## 5. Callbacks\n",
    "\n",
    "nestkit supports a **callback system** that hooks into five lifecycle events:\n",
    "\n",
    "1. `on_outer_fold_start`  -  before inner search begins\n",
    "2. `on_inner_search_complete`  -  after hyperparameter tuning\n",
    "3. `on_post_processing_complete`  -  after calibration / threshold optimization\n",
    "4. `on_outer_fold_complete`  -  after outer fold evaluation\n",
    "5. `on_nested_cv_complete`  -  after all folds finish\n",
    "\n",
    "Three built-in callbacks are provided:\n",
    "\n",
    "| Callback | Purpose |\n",
    "|----------|---------|\n",
    "| `ProgressCallback` | tqdm progress bar |\n",
    "| `LoggingCallback` | Structured log messages with timing |\n",
    "| `CheckpointCallback` | Pickle fold results to disk after each fold |\n",
    "\n",
    "Multiple callbacks can be combined by passing a list."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 28,
   "id": "a3ibgnlbhko",
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Outer folds: 100%|██████████| 5/5 [00:03<00:00,  1.36it/s]"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Done!\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\n"
     ]
    }
   ],
   "source": [
    "# ProgressCallback  -  tqdm progress bar\n",
    "ncv_progress = NestedCVClassifier(\n",
    "    estimator=RandomForestClassifier(random_state=42),\n",
    "    param_grid={\"n_estimators\": [50, 100], \"max_depth\": [3, 5]},\n",
    "    outer_cv=5,\n",
    "    inner_cv=3,\n",
    "    scoring=\"roc_auc\",\n",
    "    random_state=42,\n",
    "    callbacks=[ProgressCallback(n_outer_folds=5)],\n",
    ")\n",
    "ncv_progress.fit(X_clf, y_clf)\n",
    "print(\"Done!\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 29,
   "id": "q1zn1ye73xb",
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "nestkit | Outer fold 0: starting\n",
      "nestkit | Fold 0: started (train=455, test=114)\n",
      "nestkit | Outer fold 0: inner search complete (0.6s), best_score=0.9897\n",
      "nestkit | Fold 0: inner search complete, best_params={'max_depth': 5, 'n_estimators': 50}, best_score=0.9897\n",
      "nestkit | Fold 0: post-processing complete\n",
      "nestkit | Fold 0: complete (0.7s)\n",
      "nestkit | Outer fold 0: complete\n",
      "nestkit | Outer fold 1: starting\n",
      "nestkit | Fold 1: started (train=455, test=114)\n",
      "nestkit | Outer fold 1: inner search complete (0.7s), best_score=0.9937\n",
      "nestkit | Fold 1: inner search complete, best_params={'max_depth': 5, 'n_estimators': 100}, best_score=0.9937\n",
      "nestkit | Fold 1: post-processing complete\n",
      "nestkit | Fold 1: complete (0.8s)\n",
      "nestkit | Outer fold 1: complete\n",
      "nestkit | Outer fold 2: starting\n",
      "nestkit | Fold 2: started (train=455, test=114)\n",
      "nestkit | Outer fold 2: inner search complete (0.7s), best_score=0.9884\n",
      "nestkit | Fold 2: inner search complete, best_params={'max_depth': 5, 'n_estimators': 50}, best_score=0.9884\n",
      "nestkit | Fold 2: post-processing complete\n",
      "nestkit | Fold 2: complete (0.7s)\n",
      "nestkit | Outer fold 2: complete\n",
      "nestkit | Outer fold 3: starting\n",
      "nestkit | Fold 3: started (train=455, test=114)\n",
      "nestkit | Outer fold 3: inner search complete (0.7s), best_score=0.9931\n",
      "nestkit | Fold 3: inner search complete, best_params={'max_depth': 5, 'n_estimators': 50}, best_score=0.9931\n",
      "nestkit | Fold 3: post-processing complete\n",
      "nestkit | Fold 3: complete (0.7s)\n",
      "nestkit | Outer fold 3: complete\n",
      "nestkit | Outer fold 4: starting\n",
      "nestkit | Fold 4: started (train=456, test=113)\n",
      "nestkit | Outer fold 4: inner search complete (0.7s), best_score=0.9908\n",
      "nestkit | Fold 4: inner search complete, best_params={'max_depth': 5, 'n_estimators': 100}, best_score=0.9908\n",
      "nestkit | Fold 4: post-processing complete\n",
      "nestkit | Fold 4: complete (0.8s)\n",
      "nestkit | Outer fold 4: complete\n",
      "nestkit | Nested CV complete: 5 folds\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",
       "  /* Specific color for light theme */\n",
       "  --sklearn-color-text-on-default-background: var(--sg-text-color, var(--theme-code-foreground, var(--jp-content-font-color1, black)));\n",
       "  --sklearn-color-background: var(--sg-background-color, var(--theme-background, var(--jp-layout-color0, white)));\n",
       "  --sklearn-color-border-box: var(--sg-text-color, var(--theme-code-foreground, var(--jp-content-font-color1, black)));\n",
       "  --sklearn-color-icon: #696969;\n",
       "\n",
       "  @media (prefers-color-scheme: dark) {\n",
       "    /* Redefinition of color scheme for dark theme */\n",
       "    --sklearn-color-text-on-default-background: var(--sg-text-color, var(--theme-code-foreground, var(--jp-content-font-color1, white)));\n",
       "    --sklearn-color-background: var(--sg-background-color, var(--theme-background, var(--jp-layout-color0, #111)));\n",
       "    --sklearn-color-border-box: var(--sg-text-color, var(--theme-code-foreground, var(--jp-content-font-color1, white)));\n",
       "    --sklearn-color-icon: #878787;\n",
       "  }\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: start;\n",
       "  justify-content: space-between;\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",
       "  display: inline-block;\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-background);\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-1) 1pt solid;\n",
       "  color: var(--sklearn-color-unfitted-level-1);\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",
       "  border: var(--sklearn-color-fitted-level-1) 1pt solid;\n",
       "  color: var(--sklearn-color-fitted-level-1);\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",
       "  color: var(--sklearn-color-background);\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",
       "  color: var(--sklearn-color-background);\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-background);\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",
       "  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 summary {\n",
       "    padding: .5rem;\n",
       "    font-family: monospace;\n",
       "    cursor: pointer;\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",
       "}\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",
       ".user-set td {\n",
       "    color:rgb(255, 94, 0);\n",
       "    text-align: left;\n",
       "}\n",
       "\n",
       ".user-set td.value pre {\n",
       "    color:rgb(255, 94, 0) !important;\n",
       "    background-color: transparent !important;\n",
       "}\n",
       "\n",
       ".default td {\n",
       "    color: black;\n",
       "    text-align: left;\n",
       "}\n",
       "\n",
       ".user-set td i,\n",
       ".default td i {\n",
       "    color: black;\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>NestedCVClassifier(callbacks=[&lt;nestkit.callbacks.LoggingCallback object at 0x7f01b058f370&gt;],\n",
       "                   estimator=RandomForestClassifier(random_state=42),\n",
       "                   inner_cv=3,\n",
       "                   param_grid={&#x27;max_depth&#x27;: [3, 5], &#x27;n_estimators&#x27;: [50, 100]},\n",
       "                   random_state=42, scoring=&#x27;roc_auc&#x27;)</pre><b>In a Jupyter environment, please rerun this cell to show the HTML representation or trust the notebook. <br />On GitHub, the HTML representation is unable to render, please try loading this page with nbviewer.org.</b></div><div class=\"sk-container\" hidden><div class=\"sk-item sk-dashed-wrapped\"><div class=\"sk-label-container\"><div class=\"sk-label fitted sk-toggleable\"><input class=\"sk-toggleable__control sk-hidden--visually\" id=\"sk-estimator-id-1\" type=\"checkbox\" ><label for=\"sk-estimator-id-1\" class=\"sk-toggleable__label fitted sk-toggleable__label-arrow\"><div><div>NestedCVClassifier</div></div><div><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=\"user-set\">\n",
       "            <td><i class=\"copy-paste-icon\"\n",
       "                 onclick=\"copyToClipboard('estimator',\n",
       "                          this.parentElement.nextElementSibling)\"\n",
       "            ></i></td>\n",
       "            <td class=\"param\">estimator&nbsp;</td>\n",
       "            <td class=\"value\">RandomForestC...ndom_state=42)</td>\n",
       "        </tr>\n",
       "    \n",
       "\n",
       "        <tr class=\"user-set\">\n",
       "            <td><i class=\"copy-paste-icon\"\n",
       "                 onclick=\"copyToClipboard('param_grid',\n",
       "                          this.parentElement.nextElementSibling)\"\n",
       "            ></i></td>\n",
       "            <td class=\"param\">param_grid&nbsp;</td>\n",
       "            <td class=\"value\">{&#x27;max_depth&#x27;: [3, 5], &#x27;n_estimators&#x27;: [50, 100]}</td>\n",
       "        </tr>\n",
       "    \n",
       "\n",
       "        <tr class=\"default\">\n",
       "            <td><i class=\"copy-paste-icon\"\n",
       "                 onclick=\"copyToClipboard('search_strategy',\n",
       "                          this.parentElement.nextElementSibling)\"\n",
       "            ></i></td>\n",
       "            <td class=\"param\">search_strategy&nbsp;</td>\n",
       "            <td class=\"value\">&#x27;grid&#x27;</td>\n",
       "        </tr>\n",
       "    \n",
       "\n",
       "        <tr class=\"default\">\n",
       "            <td><i class=\"copy-paste-icon\"\n",
       "                 onclick=\"copyToClipboard('outer_cv',\n",
       "                          this.parentElement.nextElementSibling)\"\n",
       "            ></i></td>\n",
       "            <td class=\"param\">outer_cv&nbsp;</td>\n",
       "            <td class=\"value\">5</td>\n",
       "        </tr>\n",
       "    \n",
       "\n",
       "        <tr class=\"user-set\">\n",
       "            <td><i class=\"copy-paste-icon\"\n",
       "                 onclick=\"copyToClipboard('inner_cv',\n",
       "                          this.parentElement.nextElementSibling)\"\n",
       "            ></i></td>\n",
       "            <td class=\"param\">inner_cv&nbsp;</td>\n",
       "            <td class=\"value\">3</td>\n",
       "        </tr>\n",
       "    \n",
       "\n",
       "        <tr class=\"user-set\">\n",
       "            <td><i class=\"copy-paste-icon\"\n",
       "                 onclick=\"copyToClipboard('scoring',\n",
       "                          this.parentElement.nextElementSibling)\"\n",
       "            ></i></td>\n",
       "            <td class=\"param\">scoring&nbsp;</td>\n",
       "            <td class=\"value\">&#x27;roc_auc&#x27;</td>\n",
       "        </tr>\n",
       "    \n",
       "\n",
       "        <tr class=\"default\">\n",
       "            <td><i class=\"copy-paste-icon\"\n",
       "                 onclick=\"copyToClipboard('refit',\n",
       "                          this.parentElement.nextElementSibling)\"\n",
       "            ></i></td>\n",
       "            <td class=\"param\">refit&nbsp;</td>\n",
       "            <td class=\"value\">True</td>\n",
       "        </tr>\n",
       "    \n",
       "\n",
       "        <tr class=\"default\">\n",
       "            <td><i class=\"copy-paste-icon\"\n",
       "                 onclick=\"copyToClipboard('return_train_score',\n",
       "                          this.parentElement.nextElementSibling)\"\n",
       "            ></i></td>\n",
       "            <td class=\"param\">return_train_score&nbsp;</td>\n",
       "            <td class=\"value\">False</td>\n",
       "        </tr>\n",
       "    \n",
       "\n",
       "        <tr class=\"default\">\n",
       "            <td><i class=\"copy-paste-icon\"\n",
       "                 onclick=\"copyToClipboard('return_estimator',\n",
       "                          this.parentElement.nextElementSibling)\"\n",
       "            ></i></td>\n",
       "            <td class=\"param\">return_estimator&nbsp;</td>\n",
       "            <td class=\"value\">True</td>\n",
       "        </tr>\n",
       "    \n",
       "\n",
       "        <tr class=\"default\">\n",
       "            <td><i class=\"copy-paste-icon\"\n",
       "                 onclick=\"copyToClipboard('error_score',\n",
       "                          this.parentElement.nextElementSibling)\"\n",
       "            ></i></td>\n",
       "            <td class=\"param\">error_score&nbsp;</td>\n",
       "            <td class=\"value\">&#x27;raise&#x27;</td>\n",
       "        </tr>\n",
       "    \n",
       "\n",
       "        <tr class=\"default\">\n",
       "            <td><i class=\"copy-paste-icon\"\n",
       "                 onclick=\"copyToClipboard('n_jobs_outer',\n",
       "                          this.parentElement.nextElementSibling)\"\n",
       "            ></i></td>\n",
       "            <td class=\"param\">n_jobs_outer&nbsp;</td>\n",
       "            <td class=\"value\">None</td>\n",
       "        </tr>\n",
       "    \n",
       "\n",
       "        <tr class=\"default\">\n",
       "            <td><i class=\"copy-paste-icon\"\n",
       "                 onclick=\"copyToClipboard('n_jobs_inner',\n",
       "                          this.parentElement.nextElementSibling)\"\n",
       "            ></i></td>\n",
       "            <td class=\"param\">n_jobs_inner&nbsp;</td>\n",
       "            <td class=\"value\">None</td>\n",
       "        </tr>\n",
       "    \n",
       "\n",
       "        <tr class=\"default\">\n",
       "            <td><i class=\"copy-paste-icon\"\n",
       "                 onclick=\"copyToClipboard('verbose',\n",
       "                          this.parentElement.nextElementSibling)\"\n",
       "            ></i></td>\n",
       "            <td class=\"param\">verbose&nbsp;</td>\n",
       "            <td class=\"value\">0</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\">random_state&nbsp;</td>\n",
       "            <td class=\"value\">42</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\">callbacks&nbsp;</td>\n",
       "            <td class=\"value\">[&lt;nestkit.call...x7f01b058f370&gt;]</td>\n",
       "        </tr>\n",
       "    \n",
       "\n",
       "        <tr class=\"default\">\n",
       "            <td><i class=\"copy-paste-icon\"\n",
       "                 onclick=\"copyToClipboard('pre_dispatch',\n",
       "                          this.parentElement.nextElementSibling)\"\n",
       "            ></i></td>\n",
       "            <td class=\"param\">pre_dispatch&nbsp;</td>\n",
       "            <td class=\"value\">&#x27;2*n_jobs&#x27;</td>\n",
       "        </tr>\n",
       "    \n",
       "\n",
       "        <tr class=\"default\">\n",
       "            <td><i class=\"copy-paste-icon\"\n",
       "                 onclick=\"copyToClipboard('calibration_method',\n",
       "                          this.parentElement.nextElementSibling)\"\n",
       "            ></i></td>\n",
       "            <td class=\"param\">calibration_method&nbsp;</td>\n",
       "            <td class=\"value\">None</td>\n",
       "        </tr>\n",
       "    \n",
       "\n",
       "        <tr class=\"default\">\n",
       "            <td><i class=\"copy-paste-icon\"\n",
       "                 onclick=\"copyToClipboard('threshold_strategy',\n",
       "                          this.parentElement.nextElementSibling)\"\n",
       "            ></i></td>\n",
       "            <td class=\"param\">threshold_strategy&nbsp;</td>\n",
       "            <td class=\"value\">None</td>\n",
       "        </tr>\n",
       "    \n",
       "\n",
       "        <tr class=\"default\">\n",
       "            <td><i class=\"copy-paste-icon\"\n",
       "                 onclick=\"copyToClipboard('threshold_criterion',\n",
       "                          this.parentElement.nextElementSibling)\"\n",
       "            ></i></td>\n",
       "            <td class=\"param\">threshold_criterion&nbsp;</td>\n",
       "            <td class=\"value\">&#x27;youden&#x27;</td>\n",
       "        </tr>\n",
       "    \n",
       "\n",
       "        <tr class=\"default\">\n",
       "            <td><i class=\"copy-paste-icon\"\n",
       "                 onclick=\"copyToClipboard('threshold_beta',\n",
       "                          this.parentElement.nextElementSibling)\"\n",
       "            ></i></td>\n",
       "            <td class=\"param\">threshold_beta&nbsp;</td>\n",
       "            <td class=\"value\">1.0</td>\n",
       "        </tr>\n",
       "    \n",
       "\n",
       "        <tr class=\"default\">\n",
       "            <td><i class=\"copy-paste-icon\"\n",
       "                 onclick=\"copyToClipboard('cost_matrix',\n",
       "                          this.parentElement.nextElementSibling)\"\n",
       "            ></i></td>\n",
       "            <td class=\"param\">cost_matrix&nbsp;</td>\n",
       "            <td class=\"value\">None</td>\n",
       "        </tr>\n",
       "    \n",
       "\n",
       "        <tr class=\"default\">\n",
       "            <td><i class=\"copy-paste-icon\"\n",
       "                 onclick=\"copyToClipboard('min_recall',\n",
       "                          this.parentElement.nextElementSibling)\"\n",
       "            ></i></td>\n",
       "            <td class=\"param\">min_recall&nbsp;</td>\n",
       "            <td class=\"value\">None</td>\n",
       "        </tr>\n",
       "    \n",
       "\n",
       "        <tr class=\"default\">\n",
       "            <td><i class=\"copy-paste-icon\"\n",
       "                 onclick=\"copyToClipboard('calibration_cv',\n",
       "                          this.parentElement.nextElementSibling)\"\n",
       "            ></i></td>\n",
       "            <td class=\"param\">calibration_cv&nbsp;</td>\n",
       "            <td class=\"value\">None</td>\n",
       "        </tr>\n",
       "    \n",
       "\n",
       "        <tr class=\"default\">\n",
       "            <td><i class=\"copy-paste-icon\"\n",
       "                 onclick=\"copyToClipboard('conformal_prediction',\n",
       "                          this.parentElement.nextElementSibling)\"\n",
       "            ></i></td>\n",
       "            <td class=\"param\">conformal_prediction&nbsp;</td>\n",
       "            <td class=\"value\">False</td>\n",
       "        </tr>\n",
       "    \n",
       "\n",
       "        <tr class=\"default\">\n",
       "            <td><i class=\"copy-paste-icon\"\n",
       "                 onclick=\"copyToClipboard('conformal_alpha',\n",
       "                          this.parentElement.nextElementSibling)\"\n",
       "            ></i></td>\n",
       "            <td class=\"param\">conformal_alpha&nbsp;</td>\n",
       "            <td class=\"value\">0.1</td>\n",
       "        </tr>\n",
       "    \n",
       "                  </tbody>\n",
       "                </table>\n",
       "            </details>\n",
       "        </div>\n",
       "    </div></div></div><div class=\"sk-parallel\"><div class=\"sk-parallel-item\"><div class=\"sk-item\"><div class=\"sk-label-container\"><div class=\"sk-label fitted sk-toggleable\"><input class=\"sk-toggleable__control sk-hidden--visually\" id=\"sk-estimator-id-2\" type=\"checkbox\" ><label for=\"sk-estimator-id-2\" class=\"sk-toggleable__label fitted sk-toggleable__label-arrow\"><div><div>estimator: RandomForestClassifier</div></div></label><div class=\"sk-toggleable__content fitted\" data-param-prefix=\"estimator__\"><pre>RandomForestClassifier(random_state=42)</pre></div></div></div><div class=\"sk-serial\"><div class=\"sk-item\"><div class=\"sk-estimator fitted sk-toggleable\"><input class=\"sk-toggleable__control sk-hidden--visually\" id=\"sk-estimator-id-3\" type=\"checkbox\" ><label for=\"sk-estimator-id-3\" class=\"sk-toggleable__label fitted sk-toggleable__label-arrow\"><div><div>RandomForestClassifier</div></div><div><a class=\"sk-estimator-doc-link fitted\" rel=\"noreferrer\" target=\"_blank\" href=\"https://scikit-learn.org/1.7/modules/generated/sklearn.ensemble.RandomForestClassifier.html\">?<span>Documentation for RandomForestClassifier</span></a></div></label><div class=\"sk-toggleable__content fitted\" data-param-prefix=\"estimator__\">\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('n_estimators',\n",
       "                          this.parentElement.nextElementSibling)\"\n",
       "            ></i></td>\n",
       "            <td class=\"param\">n_estimators&nbsp;</td>\n",
       "            <td class=\"value\">100</td>\n",
       "        </tr>\n",
       "    \n",
       "\n",
       "        <tr class=\"default\">\n",
       "            <td><i class=\"copy-paste-icon\"\n",
       "                 onclick=\"copyToClipboard('criterion',\n",
       "                          this.parentElement.nextElementSibling)\"\n",
       "            ></i></td>\n",
       "            <td class=\"param\">criterion&nbsp;</td>\n",
       "            <td class=\"value\">&#x27;gini&#x27;</td>\n",
       "        </tr>\n",
       "    \n",
       "\n",
       "        <tr class=\"default\">\n",
       "            <td><i class=\"copy-paste-icon\"\n",
       "                 onclick=\"copyToClipboard('max_depth',\n",
       "                          this.parentElement.nextElementSibling)\"\n",
       "            ></i></td>\n",
       "            <td class=\"param\">max_depth&nbsp;</td>\n",
       "            <td class=\"value\">None</td>\n",
       "        </tr>\n",
       "    \n",
       "\n",
       "        <tr class=\"default\">\n",
       "            <td><i class=\"copy-paste-icon\"\n",
       "                 onclick=\"copyToClipboard('min_samples_split',\n",
       "                          this.parentElement.nextElementSibling)\"\n",
       "            ></i></td>\n",
       "            <td class=\"param\">min_samples_split&nbsp;</td>\n",
       "            <td class=\"value\">2</td>\n",
       "        </tr>\n",
       "    \n",
       "\n",
       "        <tr class=\"default\">\n",
       "            <td><i class=\"copy-paste-icon\"\n",
       "                 onclick=\"copyToClipboard('min_samples_leaf',\n",
       "                          this.parentElement.nextElementSibling)\"\n",
       "            ></i></td>\n",
       "            <td class=\"param\">min_samples_leaf&nbsp;</td>\n",
       "            <td class=\"value\">1</td>\n",
       "        </tr>\n",
       "    \n",
       "\n",
       "        <tr class=\"default\">\n",
       "            <td><i class=\"copy-paste-icon\"\n",
       "                 onclick=\"copyToClipboard('min_weight_fraction_leaf',\n",
       "                          this.parentElement.nextElementSibling)\"\n",
       "            ></i></td>\n",
       "            <td class=\"param\">min_weight_fraction_leaf&nbsp;</td>\n",
       "            <td class=\"value\">0.0</td>\n",
       "        </tr>\n",
       "    \n",
       "\n",
       "        <tr class=\"default\">\n",
       "            <td><i class=\"copy-paste-icon\"\n",
       "                 onclick=\"copyToClipboard('max_features',\n",
       "                          this.parentElement.nextElementSibling)\"\n",
       "            ></i></td>\n",
       "            <td class=\"param\">max_features&nbsp;</td>\n",
       "            <td class=\"value\">&#x27;sqrt&#x27;</td>\n",
       "        </tr>\n",
       "    \n",
       "\n",
       "        <tr class=\"default\">\n",
       "            <td><i class=\"copy-paste-icon\"\n",
       "                 onclick=\"copyToClipboard('max_leaf_nodes',\n",
       "                          this.parentElement.nextElementSibling)\"\n",
       "            ></i></td>\n",
       "            <td class=\"param\">max_leaf_nodes&nbsp;</td>\n",
       "            <td class=\"value\">None</td>\n",
       "        </tr>\n",
       "    \n",
       "\n",
       "        <tr class=\"default\">\n",
       "            <td><i class=\"copy-paste-icon\"\n",
       "                 onclick=\"copyToClipboard('min_impurity_decrease',\n",
       "                          this.parentElement.nextElementSibling)\"\n",
       "            ></i></td>\n",
       "            <td class=\"param\">min_impurity_decrease&nbsp;</td>\n",
       "            <td class=\"value\">0.0</td>\n",
       "        </tr>\n",
       "    \n",
       "\n",
       "        <tr class=\"default\">\n",
       "            <td><i class=\"copy-paste-icon\"\n",
       "                 onclick=\"copyToClipboard('bootstrap',\n",
       "                          this.parentElement.nextElementSibling)\"\n",
       "            ></i></td>\n",
       "            <td class=\"param\">bootstrap&nbsp;</td>\n",
       "            <td class=\"value\">True</td>\n",
       "        </tr>\n",
       "    \n",
       "\n",
       "        <tr class=\"default\">\n",
       "            <td><i class=\"copy-paste-icon\"\n",
       "                 onclick=\"copyToClipboard('oob_score',\n",
       "                          this.parentElement.nextElementSibling)\"\n",
       "            ></i></td>\n",
       "            <td class=\"param\">oob_score&nbsp;</td>\n",
       "            <td class=\"value\">False</td>\n",
       "        </tr>\n",
       "    \n",
       "\n",
       "        <tr class=\"default\">\n",
       "            <td><i class=\"copy-paste-icon\"\n",
       "                 onclick=\"copyToClipboard('n_jobs',\n",
       "                          this.parentElement.nextElementSibling)\"\n",
       "            ></i></td>\n",
       "            <td class=\"param\">n_jobs&nbsp;</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\">random_state&nbsp;</td>\n",
       "            <td class=\"value\">42</td>\n",
       "        </tr>\n",
       "    \n",
       "\n",
       "        <tr class=\"default\">\n",
       "            <td><i class=\"copy-paste-icon\"\n",
       "                 onclick=\"copyToClipboard('verbose',\n",
       "                          this.parentElement.nextElementSibling)\"\n",
       "            ></i></td>\n",
       "            <td class=\"param\">verbose&nbsp;</td>\n",
       "            <td class=\"value\">0</td>\n",
       "        </tr>\n",
       "    \n",
       "\n",
       "        <tr class=\"default\">\n",
       "            <td><i class=\"copy-paste-icon\"\n",
       "                 onclick=\"copyToClipboard('warm_start',\n",
       "                          this.parentElement.nextElementSibling)\"\n",
       "            ></i></td>\n",
       "            <td class=\"param\">warm_start&nbsp;</td>\n",
       "            <td class=\"value\">False</td>\n",
       "        </tr>\n",
       "    \n",
       "\n",
       "        <tr class=\"default\">\n",
       "            <td><i class=\"copy-paste-icon\"\n",
       "                 onclick=\"copyToClipboard('class_weight',\n",
       "                          this.parentElement.nextElementSibling)\"\n",
       "            ></i></td>\n",
       "            <td class=\"param\">class_weight&nbsp;</td>\n",
       "            <td class=\"value\">None</td>\n",
       "        </tr>\n",
       "    \n",
       "\n",
       "        <tr class=\"default\">\n",
       "            <td><i class=\"copy-paste-icon\"\n",
       "                 onclick=\"copyToClipboard('ccp_alpha',\n",
       "                          this.parentElement.nextElementSibling)\"\n",
       "            ></i></td>\n",
       "            <td class=\"param\">ccp_alpha&nbsp;</td>\n",
       "            <td class=\"value\">0.0</td>\n",
       "        </tr>\n",
       "    \n",
       "\n",
       "        <tr class=\"default\">\n",
       "            <td><i class=\"copy-paste-icon\"\n",
       "                 onclick=\"copyToClipboard('max_samples',\n",
       "                          this.parentElement.nextElementSibling)\"\n",
       "            ></i></td>\n",
       "            <td class=\"param\">max_samples&nbsp;</td>\n",
       "            <td class=\"value\">None</td>\n",
       "        </tr>\n",
       "    \n",
       "\n",
       "        <tr class=\"default\">\n",
       "            <td><i class=\"copy-paste-icon\"\n",
       "                 onclick=\"copyToClipboard('monotonic_cst',\n",
       "                          this.parentElement.nextElementSibling)\"\n",
       "            ></i></td>\n",
       "            <td class=\"param\">monotonic_cst&nbsp;</td>\n",
       "            <td class=\"value\">None</td>\n",
       "        </tr>\n",
       "    \n",
       "                  </tbody>\n",
       "                </table>\n",
       "            </details>\n",
       "        </div>\n",
       "    </div></div></div></div></div></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('.fa-regular.fa-copy').forEach(function(element) {\n",
       "    const toggleableContent = element.closest('.sk-toggleable__content');\n",
       "    const paramPrefix = toggleableContent ? toggleableContent.dataset.paramPrefix : '';\n",
       "    const paramName = element.parentElement.nextElementSibling.textContent.trim();\n",
       "    const fullParamName = paramPrefix ? `${paramPrefix}${paramName}` : paramName;\n",
       "\n",
       "    element.setAttribute('title', fullParamName);\n",
       "});\n",
       "</script></body>"
      ],
      "text/plain": [
       "NestedCVClassifier(callbacks=[<nestkit.callbacks.LoggingCallback object at 0x7f01b058f370>],\n",
       "                   estimator=RandomForestClassifier(random_state=42),\n",
       "                   inner_cv=3,\n",
       "                   param_grid={'max_depth': [3, 5], 'n_estimators': [50, 100]},\n",
       "                   random_state=42, scoring='roc_auc')"
      ]
     },
     "execution_count": 29,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# LoggingCallback  -  structured log messages\n",
    "logging.basicConfig(level=logging.INFO, format=\"%(name)s | %(message)s\", force=True)\n",
    "\n",
    "ncv_logged = NestedCVClassifier(\n",
    "    estimator=RandomForestClassifier(random_state=42),\n",
    "    param_grid={\"n_estimators\": [50, 100], \"max_depth\": [3, 5]},\n",
    "    outer_cv=5,\n",
    "    inner_cv=3,\n",
    "    scoring=\"roc_auc\",\n",
    "    random_state=42,\n",
    "    callbacks=[LoggingCallback()],\n",
    ")\n",
    "ncv_logged.fit(X_clf, y_clf)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 30,
   "id": "6xaycdc2z35",
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "nestkit | Outer fold 0: starting\n",
      "nestkit | Outer fold 0: inner search complete (0.7s), best_score=0.9897\n",
      "nestkit | Checkpointed fold 0 to /tmp/tmpi03bx2tr/ncv_checkpoints/fold_0.pkl\n",
      "nestkit | Outer fold 0: complete\n",
      "nestkit | Outer fold 1: starting\n",
      "nestkit | Outer fold 1: inner search complete (0.7s), best_score=0.9937\n",
      "nestkit | Checkpointed fold 1 to /tmp/tmpi03bx2tr/ncv_checkpoints/fold_1.pkl\n",
      "nestkit | Outer fold 1: complete\n",
      "nestkit | Outer fold 2: starting\n",
      "nestkit | Outer fold 2: inner search complete (0.7s), best_score=0.9884\n",
      "nestkit | Checkpointed fold 2 to /tmp/tmpi03bx2tr/ncv_checkpoints/fold_2.pkl\n",
      "nestkit | Outer fold 2: complete\n",
      "nestkit | Outer fold 3: starting\n",
      "nestkit | Outer fold 3: inner search complete (0.7s), best_score=0.9931\n",
      "nestkit | Checkpointed fold 3 to /tmp/tmpi03bx2tr/ncv_checkpoints/fold_3.pkl\n",
      "nestkit | Outer fold 3: complete\n",
      "nestkit | Outer fold 4: starting\n",
      "nestkit | Outer fold 4: inner search complete (0.7s), best_score=0.9908\n",
      "nestkit | Checkpointed fold 4 to /tmp/tmpi03bx2tr/ncv_checkpoints/fold_4.pkl\n",
      "nestkit | Outer fold 4: complete\n",
      "nestkit | Checkpointed final results to /tmp/tmpi03bx2tr/ncv_checkpoints/final_results.pkl\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Checkpoint files: ['final_results.pkl', 'fold_0.pkl', 'fold_1.pkl', 'fold_2.pkl', 'fold_3.pkl', 'fold_4.pkl']\n"
     ]
    }
   ],
   "source": [
    "# CheckpointCallback  -  saves fold results to disk\n",
    "with tempfile.TemporaryDirectory() as tmpdir:\n",
    "    checkpoint_dir = os.path.join(tmpdir, \"ncv_checkpoints\")\n",
    "\n",
    "    ncv_ckpt = NestedCVClassifier(\n",
    "        estimator=RandomForestClassifier(random_state=42),\n",
    "        param_grid={\"n_estimators\": [50, 100], \"max_depth\": [3, 5]},\n",
    "        outer_cv=5,\n",
    "        inner_cv=3,\n",
    "        scoring=\"roc_auc\",\n",
    "        random_state=42,\n",
    "        callbacks=[CheckpointCallback(checkpoint_dir)],\n",
    "    )\n",
    "    ncv_ckpt.fit(X_clf, y_clf)\n",
    "\n",
    "    saved_files = sorted(os.listdir(checkpoint_dir))\n",
    "    print(f\"Checkpoint files: {saved_files}\")"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "w9gx06piu8",
   "metadata": {},
   "source": [
    "### 5.1 Writing a custom callback\n",
    "\n",
    "Any object that implements the five hook methods of `FoldCallback` can be used as a callback. Here's a minimal example that records per-fold wall-clock time:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 31,
   "id": "xgqa1twby5",
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "nestkit | Outer fold 0: starting\n",
      "nestkit | Outer fold 0: inner search complete (0.6s), best_score=0.9897\n",
      "nestkit | Outer fold 0: complete\n",
      "nestkit | Outer fold 1: starting\n",
      "nestkit | Outer fold 1: inner search complete (0.7s), best_score=0.9937\n",
      "nestkit | Outer fold 1: complete\n",
      "nestkit | Outer fold 2: starting\n",
      "nestkit | Outer fold 2: inner search complete (0.7s), best_score=0.9884\n",
      "nestkit | Outer fold 2: complete\n",
      "nestkit | Outer fold 3: starting\n",
      "nestkit | Outer fold 3: inner search complete (0.7s), best_score=0.9931\n",
      "nestkit | Outer fold 3: complete\n",
      "nestkit | Outer fold 4: starting\n",
      "nestkit | Outer fold 4: inner search complete (0.7s), best_score=0.9908\n",
      "nestkit | Outer fold 4: complete\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Per-fold wall-clock time:\n",
      "  Fold 0: 0.69s\n",
      "  Fold 1: 0.77s\n",
      "  Fold 2: 0.70s\n",
      "  Fold 3: 0.72s\n",
      "  Fold 4: 0.82s\n"
     ]
    }
   ],
   "source": [
    "class TimingCallback:\n",
    "    \"\"\"Record wall-clock time per outer fold.\"\"\"\n",
    "\n",
    "    def __init__(self):\n",
    "        self.fold_times = {}\n",
    "        self._start = None\n",
    "\n",
    "    def on_outer_fold_start(self, fold_idx, train_idx, test_idx):\n",
    "        self._start = time.time()\n",
    "\n",
    "    def on_inner_search_complete(self, fold_idx, search):\n",
    "        pass\n",
    "\n",
    "    def on_post_processing_complete(self, fold_idx, artifacts):\n",
    "        pass\n",
    "\n",
    "    def on_outer_fold_complete(self, fold_idx, result):\n",
    "        self.fold_times[fold_idx] = time.time() - self._start\n",
    "\n",
    "    def on_nested_cv_complete(self, results):\n",
    "        pass\n",
    "\n",
    "\n",
    "timer = TimingCallback()\n",
    "ncv_timed = NestedCVClassifier(\n",
    "    estimator=RandomForestClassifier(random_state=42),\n",
    "    param_grid={\"n_estimators\": [50, 100], \"max_depth\": [3, 5]},\n",
    "    outer_cv=5,\n",
    "    inner_cv=3,\n",
    "    scoring=\"roc_auc\",\n",
    "    random_state=42,\n",
    "    callbacks=[timer],\n",
    ")\n",
    "ncv_timed.fit(X_clf, y_clf)\n",
    "\n",
    "print(\"Per-fold wall-clock time:\")\n",
    "for fold, elapsed in timer.fold_times.items():\n",
    "    print(f\"  Fold {fold}: {elapsed:.2f}s\")"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "qzta5vx853f",
   "metadata": {},
   "source": [
    "## Summary\n",
    "\n",
    "This notebook demonstrated nestkit's advanced analysis tools:\n",
    "\n",
    "| Tool | Purpose |\n",
    "|------|--------|\n",
    "| `NestedCVComparator` | Statistically rigorous model comparison (corrected t-tests, Bayesian tests, critical difference diagrams) |\n",
    "| `FeatureImportanceAggregator` | Stable, cross-fold feature importance with Nogueira stability index |\n",
    "| `HyperparameterStability` | Diagnose hyperparameter selection consistency |\n",
    "| Conformal prediction | CV+ Mondrian prediction sets (classifier) and conditional intervals (regressor) |\n",
    "| Callbacks | Monitor, log, and checkpoint nested CV runs |\n",
    "\n",
    "### Key takeaways\n",
    "\n",
    "1. **Never compare models on mean scores alone**  -  use corrected statistical tests or Bayesian comparisons\n",
    "2. **Feature importances from a single split are unreliable**  -  aggregate across folds and check stability\n",
    "3. **Unstable hyperparameters are a warning sign**  -  high entropy or low Jaccard similarity suggests sensitivity to data composition\n",
    "4. **Conformal prediction quantifies uncertainty**  -  prediction sets (classification) and conditional intervals (regression) provide coverage guarantees\n",
    "5. **Callbacks make long runs observable**  -  combine progress bars, logging, and checkpointing"
   ]
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "nestkit",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.10.12"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 5
}