Add residual estimator and erase previous one

examples/example_advanced.ipynb

@@ -0,0 +1,143 @@
+{
+ "cells": [
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "# Advanced functionalities\n",
+    "\n",
+    "In the example notebook, we learned how to measure the grouping loss and grouping risks with partitioning estimates fitted on the residuals. We can also use broader models such as **Histogram Gradient Boosting** or **Random Forests** directly fitted on the residuals to provide estimates. \n",
+    "\n",
+    "\n",
+    "> **⚠️ Warning:** These estimates come with no theoretical guarantee in our work."
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 1,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "from sklearn.datasets import make_classification\n",
+    "from sklearn.linear_model import LogisticRegression\n",
+    "from sklearn.model_selection import train_test_split\n",
+    "\n",
+    "X, y = make_classification(\n",
+    "    n_samples=200000, random_state=42, n_features=20, n_informative=20, n_redundant=0\n",
+    ")\n",
+    "\n",
+    "X_train, X_test, y_train, y_test = train_test_split(X, y, test_size=0.5, random_state=0)\n",
+    "\n",
+    "est = LogisticRegression()\n",
+    "est.fit(X_train, y_train)\n",
+    "S_test = est.predict_proba(X_test)[:, 1]"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "To indicate that we aren't using a partitioning estimator, we use the `residual_estimator` option."
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 2,
+   "metadata": {},
+   "outputs": [
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "GLEstimator()\n",
+      "  Scoring Rule      : Brier: 0.1403\n",
+      "  Grouping loss     : 0.1313\n",
+      "  Calibration Loss  : 0.0025\n",
+      "  Epistemic Loss    : 0.1338\n",
+      "\n"
+     ]
+    }
+   ],
+   "source": [
+    "from glest.core import GLEstimator\n",
+    "from sklearn.ensemble import RandomForestRegressor\n",
+    "\n",
+    "\n",
+    "gle_custom = GLEstimator(\n",
+    "    residual_estimator=RandomForestRegressor(n_estimators=100, random_state=42)\n",
+    ")\n",
+    "gle_custom.fit(X_test, S_test, y_test)\n",
+    "gle_custom.estimate()\n",
+    "\n",
+    "print(gle_custom)"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "The same goes for the RiskEstimator"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 3,
+   "metadata": {},
+   "outputs": [
+    {
+     "data": {
+      "text/plain": [
+       "(array([0., 0., 0., ..., 0., 0., 0.], shape=(50000,)),\n",
+       " array([0., 0., 0., ..., 0., 0., 0.], shape=(50000,)))"
+      ]
+     },
+     "execution_count": 3,
+     "metadata": {},
+     "output_type": "execute_result"
+    }
+   ],
+   "source": [
+    "from glest.core import RiskEstimator\n",
+    "\n",
+    "X_test, X_risk, y_test, y_risk = train_test_split(\n",
+    "    X_test, y_test, test_size=0.5, random_state=0\n",
+    ")\n",
+    "\n",
+    "\n",
+    "S_test = est.predict_proba(X_test)[:, 1]\n",
+    "\n",
+    "risk = RiskEstimator(\n",
+    "    residual_estimator=RandomForestRegressor(n_estimators=100, random_state=42)\n",
+    ")\n",
+    "risk.fit(X_test, S_test, y_test)\n",
+    "risk.predict(X_risk, est.predict_proba(X_risk)[:, 1].reshape(-1, 1), t=0.5)"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": []
+  }
+ ],
+ "metadata": {
+  "kernelspec": {
+   "display_name": "glestest",
+   "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": "3.11.13"
+  }
+ },
+ "nbformat": 4,
+ "nbformat_minor": 2
+}

glest/__init__.py

@@ -1,3 +1,3 @@
-__version__ = "0.0.1-alpha.1"
+__version__ = "0.0.2-alpha.1"
 
-from .core import GLEstimator, GLEstimatorCV, Partitioner
+from .core import GLEstimator, PartitioningEstimate, RiskEstimator