{ "cells": [ { "cell_type": "markdown", "id": "52dfcfb3", "metadata": {}, "source": [ "# Ensembles of GUIDE trees\n", "\n", "guide-rs fits three kinds of tree ensembles, all of which choose split variables with GUIDE's chi-square tests rather than by exhaustive search:\n", "\n", "- **Bagging** (`GuideForestRegressor(kind=\"bagging\")`, `GuideForestClassifier(kind=\"bagging\")`): unpruned GUIDE trees fitted to bootstrap samples, with every variable a candidate at every node.\n", "- **Forests** (`GuideForestRegressor`, `GuideForestClassifier`): bagging that also samples the candidate split variables at each node, a third of them for regression and the square root of their number for classification.\n", "- **Boosting** (`GuideBoostingRegressor`, `GuideBoostingClassifier`): shallow trees fitted in sequence to the gradients of the loss. chi-square tests of the gradients against each predictor choose each split variable (GUIDE's tests on their signs for squared error, score tests for bin effects for classification), the second-order (Newton) gain chooses the split point, and each leaf holds a shrunken Newton step. Boosting is an experimental guide-rs extension; GUIDE itself does not boost, and the selection tests and defaults may change.\n", "\n", "This notebook compares them with a single GUIDE tree and with scikit-learn's `BaggingRegressor`, `RandomForestRegressor`/`RandomForestClassifier` and `HistGradientBoostingRegressor`/`HistGradientBoostingClassifier`. The selection bias notebook shows how GUIDE boosting spreads its splits evenly over pure-noise predictors." ] }, { "cell_type": "code", "execution_count": 1, "id": "d9a15aa8", "metadata": { "execution": { "iopub.execute_input": "2026-11-02T22:20:15.677845Z", "iopub.status.busy": "2026-11-02T22:20:15.677496Z", "iopub.status.idle": "2026-11-02T22:20:17.353822Z", "shell.execute_reply": "2026-11-02T22:20:17.352999Z" } }, "outputs": [], "source": [ "import json\n", "\n", "import matplotlib.pyplot as plt\n", "import numpy as np\n", "import pandas as pd\n", "from sklearn.calibration import calibration_curve\n", "from sklearn.datasets import load_breast_cancer, load_diabetes, load_wine, make_friedman1\n", "from sklearn.ensemble import (\n", " BaggingRegressor,\n", " HistGradientBoostingClassifier,\n", " HistGradientBoostingRegressor,\n", " RandomForestClassifier,\n", " RandomForestRegressor,\n", ")\n", "from sklearn.model_selection import (\n", " KFold,\n", " StratifiedKFold,\n", " cross_val_predict,\n", " cross_validate,\n", " train_test_split,\n", ")\n", "from sklearn.tree import DecisionTreeRegressor\n", "\n", "from guide import (\n", " GuideBoostingClassifier,\n", " GuideBoostingRegressor,\n", " GuideClassifier,\n", " GuideForestClassifier,\n", " GuideForestRegressor,\n", " GuideRegressor,\n", " plot_tree,\n", ")\n", "\n", "pd.set_option(\"display.width\", 120)\n", "\n", "\n", "def compare(models, X, y, cv, scoring):\n", " # One column per scorer; scikit-learn's negated losses (\"neg_...\") are\n", " # reported as losses.\n", " rows = []\n", " for name, model in models.items():\n", " result = cross_validate(model, X, y, cv=cv, scoring=scoring)\n", " row = {\"model\": name}\n", " for metric, scorer in scoring.items():\n", " loss = getattr(scorer, \"negated\", False) or str(scorer).startswith(\"neg_\")\n", " row[metric] = (-1 if loss else 1) * result[f\"test_{metric}\"].mean()\n", " row[\"fit seconds\"] = result[\"fit_time\"].mean()\n", " rows.append(row)\n", " return pd.DataFrame(rows).set_index(\"model\").round(3)" ] }, { "cell_type": "markdown", "id": "8d27d0f8", "metadata": {}, "source": [ "## Regression\n", "\n", "Two datasets: scikit-learn's diabetes data (442 patients, ten baseline variables), and 1000 draws from Friedman's first benchmark function, $y = 10 \\sin(\\pi x_1 x_2) + 20 (x_3 - 0.5)^2 + 10 x_4 + 5 x_5 + \\varepsilon$, with five further predictors that are pure noise. Every ensemble has 100 trees except the boosted ones, which take 200 rounds at learning rate 0.05 and depth 3. We report 5-fold cross-validated R² and mean fit time; all ensembles fit on all cores." ] }, { "cell_type": "code", "execution_count": 2, "id": "ec048561", "metadata": { "execution": { "iopub.execute_input": "2026-11-02T22:20:17.355736Z", "iopub.status.busy": "2026-11-02T22:20:17.355543Z", "iopub.status.idle": "2026-11-02T22:20:31.446212Z", "shell.execute_reply": "2026-11-02T22:20:31.445697Z" } }, "outputs": [ { "data": { "text/html": [ "
\n", "\n", "\n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", "
R²fit seconds
model
diabetesGUIDE tree0.3410.020
GUIDE bagging0.4430.099
GUIDE forest0.4470.046
GUIDE boosting0.4310.201
CART bagging0.4190.407
random forest0.4420.035
HistGradientBoosting0.3950.375
Friedman #1GUIDE tree0.7080.038
GUIDE bagging0.8420.186
GUIDE forest0.8110.124
GUIDE boosting0.9110.267
CART bagging0.8490.061
random forest0.8360.038
HistGradientBoosting0.9190.782
\n", "
" ], "text/plain": [ " R² fit seconds\n", " model \n", "diabetes GUIDE tree 0.341 0.020\n", " GUIDE bagging 0.443 0.099\n", " GUIDE forest 0.447 0.046\n", " GUIDE boosting 0.431 0.201\n", " CART bagging 0.419 0.407\n", " random forest 0.442 0.035\n", " HistGradientBoosting 0.395 0.375\n", "Friedman #1 GUIDE tree 0.708 0.038\n", " GUIDE bagging 0.842 0.186\n", " GUIDE forest 0.811 0.124\n", " GUIDE boosting 0.911 0.267\n", " CART bagging 0.849 0.061\n", " random forest 0.836 0.038\n", " HistGradientBoosting 0.919 0.782" ] }, "execution_count": 2, "metadata": {}, "output_type": "execute_result" } ], "source": [ "X_diabetes, y_diabetes = load_diabetes(return_X_y=True, as_frame=True)\n", "X_friedman, y_friedman = make_friedman1(n_samples=1000, noise=1.0, random_state=0)\n", "X_friedman = pd.DataFrame(X_friedman, columns=[f\"x{i}\" for i in range(1, 11)])\n", "\n", "regressors = {\n", " \"GUIDE tree\": GuideRegressor(),\n", " \"GUIDE bagging\": GuideForestRegressor(kind=\"bagging\"),\n", " \"GUIDE forest\": GuideForestRegressor(),\n", " \"GUIDE boosting\": GuideBoostingRegressor(n_estimators=200, learning_rate=0.05),\n", " \"CART bagging\": BaggingRegressor(DecisionTreeRegressor(), n_estimators=100, n_jobs=-1,\n", " random_state=0),\n", " \"random forest\": RandomForestRegressor(n_estimators=100, max_features=1 / 3, n_jobs=-1,\n", " random_state=0),\n", " \"HistGradientBoosting\": HistGradientBoostingRegressor(max_iter=200, learning_rate=0.05,\n", " random_state=0),\n", "}\n", "cv = KFold(5, shuffle=True, random_state=0)\n", "pd.concat({\n", " \"diabetes\": compare(regressors, X_diabetes, y_diabetes, cv, {\"R²\": \"r2\"}),\n", " \"Friedman #1\": compare(regressors, X_friedman, y_friedman, cv, {\"R²\": \"r2\"}),\n", "})" ] }, { "cell_type": "markdown", "id": "1bb22b4d", "metadata": {}, "source": [ "Every ensemble improves on the single tree. On the diabetes data the gains are about 0.1 in R² and the ensembles are within 0.03 of each other, except HistGradientBoosting at these settings. Friedman's function is smooth and includes an interaction, which boosting captures much better than averaging deep trees: both boosted models reach an R² above 0.91, against 0.81 to 0.85 for bagging and forests. GUIDE boosting fits faster than HistGradientBoosting on these small datasets; HistGradientBoosting bins each feature into a histogram, which pays off on far larger data, and guide-rs does not bin yet.\n", "\n", "Bagged and forest models also report their out-of-bag loss, the mean squared error of each training row predicted by the trees whose bootstrap sample left it out. It estimates the test error without a separate cross-validation:" ] }, { "cell_type": "code", "execution_count": 3, "id": "2c5466d3", "metadata": { "execution": { "iopub.execute_input": "2026-11-02T22:20:31.447990Z", "iopub.status.busy": "2026-11-02T22:20:31.447892Z", "iopub.status.idle": "2026-11-02T22:20:31.513130Z", "shell.execute_reply": "2026-11-02T22:20:31.512560Z" } }, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "out-of-bag R²: 0.449\n" ] } ], "source": [ "forest = GuideForestRegressor().fit(X_diabetes, y_diabetes)\n", "print(f\"out-of-bag R²: {1 - forest.oob_loss_ / np.var(y_diabetes):.3f}\")" ] }, { "cell_type": "markdown", "id": "41bc45a3", "metadata": {}, "source": [ "## Early stopping\n", "\n", "Boosting adds trees until `n_estimators` is reached, so too many rounds overfit. With `early_stopping_rounds`, `fit` takes `validation_data` and stops once the validation loss has not improved for that many rounds, keeping the best round. The loss after every round is stored in the saved model:" ] }, { "cell_type": "code", "execution_count": 4, "id": "4abf9892", "metadata": { "execution": { "iopub.execute_input": "2026-11-02T22:20:31.514629Z", "iopub.status.busy": "2026-11-02T22:20:31.514527Z", "iopub.status.idle": "2026-11-02T22:20:32.348215Z", "shell.execute_reply": "2026-11-02T22:20:32.347748Z" } }, "outputs": [ { "data": { "image/png": "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", "text/plain": [ "
" ] }, "metadata": {}, "output_type": "display_data" } ], "source": [ "X_train, X_valid, y_train, y_valid = train_test_split(X_friedman, y_friedman, test_size=0.25,\n", " random_state=0)\n", "boosted = GuideBoostingRegressor(n_estimators=2000, learning_rate=0.1, early_stopping_rounds=50)\n", "boosted.fit(X_train, y_train, validation_data=(X_valid, y_valid))\n", "validation_loss = json.loads(boosted.to_json())[\"boosting\"][\"validation_loss\"]\n", "\n", "fig, ax = plt.subplots(figsize=(7, 4))\n", "ax.plot(np.arange(1, len(validation_loss) + 1), validation_loss)\n", "ax.axvline(boosted.n_estimators_, color=\"0.4\", ls=\"--\", lw=1,\n", " label=f\"best round ({boosted.n_estimators_})\")\n", "ax.set(xlabel=\"boosting round\", ylabel=\"validation loss\",\n", " title=\"GUIDE boosting on Friedman #1\")\n", "ax.legend()\n", "plt.show()" ] }, { "cell_type": "markdown", "id": "699ca108", "metadata": {}, "source": [ "The loss levels off after about 150 rounds; the best of the 377 rounds fitted is round 327, so the model keeps 327 trees. Each boosted tree is small, and its leaves hold additive scores rather than predictions. `plot_tree` draws any one of them; `apply` returns the leaf of every row in every tree, one column per tree." ] }, { "cell_type": "code", "execution_count": 5, "id": "62e06803", "metadata": { "execution": { "iopub.execute_input": "2026-11-02T22:20:32.349380Z", "iopub.status.busy": "2026-11-02T22:20:32.349317Z", "iopub.status.idle": "2026-11-02T22:20:32.472592Z", "shell.execute_reply": "2026-11-02T22:20:32.472201Z" } }, "outputs": [ { "data": { "image/png": "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", "text/plain": [ "
" ] }, "metadata": {}, "output_type": "display_data" }, { "name": "stdout", "output_type": "stream", "text": [ "apply: (250, 327)\n" ] } ], "source": [ "fig, ax = plt.subplots(figsize=(12, 4))\n", "plot_tree(boosted, tree=0, ax=ax)\n", "ax.set_title(\"First boosted tree\")\n", "plt.show()\n", "print(f\"apply: {boosted.apply(X_valid).shape}\")" ] }, { "cell_type": "markdown", "id": "5e7d90a3", "metadata": {}, "source": [ "## Classification\n", "\n", "The wine (three classes) and breast cancer (two classes) data, with the proper scoring rules from the classification notebook: log loss, the Brier score summed over classes, and AUROC (one class against the rest, averaged), plus accuracy for reference. Boosted classifiers fit one tree per round for two classes and one tree per class per round otherwise. 5-fold stratified cross-validation." ] }, { "cell_type": "code", "execution_count": 6, "id": "6643792a", "metadata": { "execution": { "iopub.execute_input": "2026-11-02T22:20:32.473664Z", "iopub.status.busy": "2026-11-02T22:20:32.473593Z", "iopub.status.idle": "2026-11-02T22:20:43.466417Z", "shell.execute_reply": "2026-11-02T22:20:43.465902Z" } }, "outputs": [ { "data": { "text/html": [ "
\n", "\n", "\n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", "
log lossBrierAUROCaccuracyfit seconds
model
wineGUIDE tree0.3920.1840.9480.8930.003
GUIDE forest0.1700.0691.0000.9890.007
GUIDE boosting0.1090.0580.9970.9660.097
random forest0.1400.0600.9990.9720.052
HistGradientBoosting0.0650.0350.9990.9770.463
breast cancerGUIDE tree0.1980.0840.9550.9560.027
GUIDE forest0.1230.0650.9910.9560.049
GUIDE boosting0.0880.0480.9930.9670.707
random forest0.1160.0620.9920.9650.042
HistGradientBoosting0.1050.0480.9940.9680.546
\n", "
" ], "text/plain": [ " log loss Brier AUROC accuracy fit seconds\n", " model \n", "wine GUIDE tree 0.392 0.184 0.948 0.893 0.003\n", " GUIDE forest 0.170 0.069 1.000 0.989 0.007\n", " GUIDE boosting 0.109 0.058 0.997 0.966 0.097\n", " random forest 0.140 0.060 0.999 0.972 0.052\n", " HistGradientBoosting 0.065 0.035 0.999 0.977 0.463\n", "breast cancer GUIDE tree 0.198 0.084 0.955 0.956 0.027\n", " GUIDE forest 0.123 0.065 0.991 0.956 0.049\n", " GUIDE boosting 0.088 0.048 0.993 0.967 0.707\n", " random forest 0.116 0.062 0.992 0.965 0.042\n", " HistGradientBoosting 0.105 0.048 0.994 0.968 0.546" ] }, "execution_count": 6, "metadata": {}, "output_type": "execute_result" } ], "source": [ "def brier(model, X, y):\n", " # Squared error of the probability vector, summed over classes (0 to 2);\n", " # negated like scikit-learn's losses.\n", " observed = np.asarray(y)[:, None] == model.classes_\n", " return -np.mean(((model.predict_proba(X) - observed) ** 2).sum(axis=1))\n", "\n", "\n", "brier.negated = True\n", "metrics = {\"log loss\": \"neg_log_loss\", \"Brier\": brier, \"AUROC\": \"roc_auc_ovr\",\n", " \"accuracy\": \"accuracy\"}\n", "\n", "\n", "def named(bunch):\n", " return bunch.data, bunch.target.map(dict(enumerate(bunch.target_names)))\n", "\n", "\n", "classifiers = {\n", " \"GUIDE tree\": GuideClassifier(),\n", " \"GUIDE forest\": GuideForestClassifier(),\n", " \"GUIDE boosting\": GuideBoostingClassifier(n_estimators=200, learning_rate=0.05),\n", " \"random forest\": RandomForestClassifier(n_estimators=100, n_jobs=-1, random_state=0),\n", " \"HistGradientBoosting\": HistGradientBoostingClassifier(max_iter=200, learning_rate=0.05,\n", " random_state=0),\n", "}\n", "datasets = {\"wine\": named(load_wine(as_frame=True)),\n", " \"breast cancer\": named(load_breast_cancer(as_frame=True))}\n", "cv5 = StratifiedKFold(5, shuffle=True, random_state=0)\n", "pd.concat({name: compare(classifiers, *data, cv5, metrics) for name, data in datasets.items()})" ] }, { "cell_type": "markdown", "id": "44330d70", "metadata": {}, "source": [ "Every ensemble's probabilities are far better than a single tree's. On breast cancer GUIDE boosting has the lowest log loss (0.087 against 0.105 for HistGradientBoosting) and ties HistGradientBoosting on the Brier score. On wine HistGradientBoosting is best, with about half the log loss of any other model, and GUIDE boosting comes next. Accuracy again ranks the models differently: on wine the GUIDE forest is the most accurate (0.989) although its log loss is the worst of the ensembles.\n", "\n", "A calibration plot compares predicted probabilities of malignancy with observed frequencies, binned by predicted probability, from the same cross-validation:" ] }, { "cell_type": "code", "execution_count": 7, "id": "62093371", "metadata": { "execution": { "iopub.execute_input": "2026-11-02T22:20:43.467976Z", "iopub.status.busy": "2026-11-02T22:20:43.467874Z", "iopub.status.idle": "2026-11-02T22:20:50.547533Z", "shell.execute_reply": "2026-11-02T22:20:50.546942Z" } }, "outputs": [ { "data": { "image/png": "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", "text/plain": [ "
" ] }, "metadata": {}, "output_type": "display_data" } ], "source": [ "X_cancer, y_cancer = datasets[\"breast cancer\"]\n", "fig, ax = plt.subplots(figsize=(5.5, 5))\n", "for name, model in classifiers.items():\n", " p = cross_val_predict(model, X_cancer, y_cancer, cv=cv5, method=\"predict_proba\")[:, 1]\n", " observed, predicted = calibration_curve(y_cancer == \"malignant\", p, n_bins=8,\n", " strategy=\"quantile\")\n", " ax.plot(predicted, observed, \"o-\", label=name)\n", "ax.plot([0, 1], [0, 1], color=\"0.6\", lw=1, ls=\"--\")\n", "ax.set(xlabel=\"predicted probability of malignant\", ylabel=\"observed frequency\",\n", " title=\"Calibration, breast cancer (5-fold CV)\")\n", "ax.legend()\n", "plt.show()" ] }, { "cell_type": "markdown", "id": "103c81f5", "metadata": {}, "source": [ "Most predictions are close to 0 or 1, so the quantile bins crowd at the ends. Both forests are underconfident at intermediate probabilities, as averages of bootstrap trees tend to be: tumors given a probability of about 0.2 are malignant about 10% of the time, and those given about 0.73 about 85% of the time. The boosted models lie close to the diagonal." ] } ], "metadata": { "kernelspec": { "display_name": "Python 3", "language": "python", "name": "python3" }, "language_info": { "codemirror_mode": { "name": "ipython", "version": 3 }, "file_extension": ".py", "mimetype": "text/x-python", "name": "python", "nbconvert_exporter": "python", "pygments_lexer": "ipython3", "version": "3.12.12" } }, "nbformat": 4, "nbformat_minor": 5 }