From 7230e5a7f3b356d678f18cb8211dff5180c89ee2 Mon Sep 17 00:00:00 2001
From: Francesco Pisu <frapisu9@gmail.com>
Date: Fri, 23 Jun 2023 12:51:46 +0200
Subject: [PATCH] docs: add usage example

---
 CHANGELOG.md       |   5 +-
 docs/example.ipynb | 409 ++++++++++++++++++++++++++++++++++++++++-----
 2 files changed, 368 insertions(+), 46 deletions(-)

diff --git a/CHANGELOG.md b/CHANGELOG.md
index 3d2342d..9b2b938 100644
--- a/CHANGELOG.md
+++ b/CHANGELOG.md
@@ -2,6 +2,7 @@
 
 <!--next-version-placeholder-->
 
-## v0.1.0 (20/06/2023)
+## v0.1.0 (23/06/2023)
 
-- First release of `modelsight`!
\ No newline at end of file
+- First release of `modelsight`!
+- Calibration module to assess calibration of ML predicted probabilities via Hosmer-Lemeshow plot.
\ No newline at end of file
diff --git a/docs/example.ipynb b/docs/example.ipynb
index 1c0b98a..69f30a7 100644
--- a/docs/example.ipynb
+++ b/docs/example.ipynb
@@ -1,45 +1,366 @@
 {
-    "cells": [
-        {
-            "cell_type": "markdown",
-            "source": [
-                "# Example usage\n",
-                "\n",
-                "To use `modelsight` in a project:"
-            ],
-            "metadata": {}
-        },
-        {
-            "cell_type": "code",
-            "execution_count": null,
-            "source": [
-                "import modelsight\n",
-                "\n",
-                "print(modelsight.__version__)"
-            ],
-            "outputs": [],
-            "metadata": {}
-        }
-    ],
-    "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": "3.8.5"
-        }
-    },
-    "nbformat": 4,
-    "nbformat_minor": 4
-}
\ No newline at end of file
+ "cells": [
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "# Usage"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "In this notebook we will see how to:\n",
+    "- train and test a binary classifier within a cross-validation framework;\n",
+    "- accumulate ground-truths and predictions and\n",
+    "- visualize calibration using the modelsight's `hosmer_lemeshow_plot` function."
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "import os\n",
+    "import sys\n",
+    "\n",
+    "sys.path.insert(0, os.path.abspath(\"../\"))"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 17,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "SEED = 1234\n",
+    "N_REPEATS = 10\n",
+    "N_SPLITS = 10"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "## Load dataset"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "In this example, we will use the Breast Cancer Wisconsin dataset, which consists of features computed from a digitized image of a fine needle aspirate of a breast mass.\n",
+    "It's a binary classification problem where the dependent variable is the biopsy's outcome: M is malignant and B is benign."
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "We can load the dataset using `scikit-learn.datasets.load_breast_cancer`."
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 4,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "from sklearn.datasets import load_breast_cancer\n",
+    "X, y = load_breast_cancer(return_X_y=True, as_frame=True)"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 6,
+   "metadata": {},
+   "outputs": [
+    {
+     "data": {
+      "text/plain": [
+       "target\n",
+       "1    0.627417\n",
+       "0    0.372583\n",
+       "Name: proportion, dtype: float64"
+      ]
+     },
+     "execution_count": 6,
+     "metadata": {},
+     "output_type": "execute_result"
+    }
+   ],
+   "source": [
+    "y.value_counts(normalize=True)"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "We can see that 62.7% of breast cancers were malignant."
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 13,
+   "metadata": {},
+   "outputs": [
+    {
+     "data": {
+      "text/plain": [
+       "mean radius                0\n",
+       "mean texture               0\n",
+       "mean perimeter             0\n",
+       "mean area                  0\n",
+       "mean smoothness            0\n",
+       "mean compactness           0\n",
+       "mean concavity             0\n",
+       "mean concave points        0\n",
+       "mean symmetry              0\n",
+       "mean fractal dimension     0\n",
+       "radius error               0\n",
+       "texture error              0\n",
+       "perimeter error            0\n",
+       "area error                 0\n",
+       "smoothness error           0\n",
+       "compactness error          0\n",
+       "concavity error            0\n",
+       "concave points error       0\n",
+       "symmetry error             0\n",
+       "fractal dimension error    0\n",
+       "worst radius               0\n",
+       "worst texture              0\n",
+       "worst perimeter            0\n",
+       "worst area                 0\n",
+       "worst smoothness           0\n",
+       "worst compactness          0\n",
+       "worst concavity            0\n",
+       "worst concave points       0\n",
+       "worst symmetry             0\n",
+       "worst fractal dimension    0\n",
+       "dtype: int64"
+      ]
+     },
+     "execution_count": 13,
+     "metadata": {},
+     "output_type": "execute_result"
+    }
+   ],
+   "source": [
+    "X.isna().sum()"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "This is a toy dataset, hence it's super clean with no missing values."
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "## Train and validate model within a cross-validation scheme"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 20,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "from interpret.glassbox import ExplainableBoostingClassifier\n",
+    "from sklearn.model_selection import RepeatedStratifiedKFold\n",
+    "\n",
+    "cv = RepeatedStratifiedKFold(n_repeats=N_REPEATS, \n",
+    "                             n_splits=N_SPLITS, \n",
+    "                             random_state=SEED)\n",
+    "\n",
+    "cv_results = {\n",
+    "    \"gt_train\": [],\n",
+    "    \"gt_val\": [],\n",
+    "    \"probas_train\": [],\n",
+    "    \"probas_val\": []\n",
+    "}\n",
+    "\n",
+    "for i, (train_idx, val_idx) in enumerate(cv.split(X, y)):\n",
+    "    Xtrain, ytrain = X.iloc[train_idx, :], y.iloc[train_idx]\n",
+    "    Xval, yval = X.iloc[val_idx, :], y.iloc[val_idx]\n",
+    "\n",
+    "    model = ExplainableBoostingClassifier(random_state=SEED,\n",
+    "                                            interactions=6,\n",
+    "                                            learning_rate=0.02,\n",
+    "                                            min_samples_leaf=5,\n",
+    "                                            n_jobs=4)\n",
+    "\n",
+    "    # this is a toy example hence we will train the model on all available features\n",
+    "    model.fit(Xtrain, ytrain)\n",
+    "\n",
+    "    # accumulate ground-truths\n",
+    "    cv_results[\"gt_train\"].append(ytrain)\n",
+    "    cv_results[\"gt_val\"].append(yval)\n",
+    "\n",
+    "    # accumulate predictions\n",
+    "    train_pred_probas = model.predict_proba(Xtrain)[:, 1]\n",
+    "    val_pred_probas = model.predict_proba(Xval)[:, 1]\n",
+    "\n",
+    "    cv_results[\"probas_train\"].append(train_pred_probas)\n",
+    "    cv_results[\"probas_val\"].append(val_pred_probas)    "
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "Let's see how this model is performing"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 23,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "import numpy as np\n",
+    "from sklearn.metrics import roc_auc_score, brier_score_loss"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 34,
+   "metadata": {},
+   "outputs": [
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "Median (95% CI) validation area under the curve: 1.00 (0.97 - 1.00)\n",
+      "Mean ± SD validation area under the curve: 0.99 ± 0.01\n"
+     ]
+    }
+   ],
+   "source": [
+    "roc_aucs = []\n",
+    "for gt, preds in zip(cv_results[\"gt_val\"], cv_results[\"probas_val\"]):\n",
+    "    roc_auc = roc_auc_score(gt, preds)\n",
+    "    roc_aucs.append(roc_auc)\n",
+    "\n",
+    "roc_low, roc_med, roc_up = np.percentile(roc_aucs, [2.5, 50, 97.5])\n",
+    "roc_mean = np.mean(roc_aucs)\n",
+    "roc_sd = np.std(roc_aucs)\n",
+    "\n",
+    "print(f\"Median (95% CI) validation area under the curve: {roc_med:.2f} ({roc_low:.2f} - {roc_up:.2f})\")\n",
+    "print(f\"Mean ± SD validation area under the curve: {roc_mean:.2f} ± {roc_sd:.2f}\")"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "We will now compute median Brier score (95% CI) and use it to annotate the calibration plot."
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 49,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "briers = []\n",
+    "for gt, preds in zip(cv_results[\"gt_val\"], cv_results[\"probas_val\"]):\n",
+    "    brier = brier_score_loss(gt, preds)\n",
+    "    briers.append(brier)\n",
+    "\n",
+    "brier_low, brier_med, brier_up = np.percentile(briers, [2.5, 50, 97.5])\n",
+    "\n",
+    "brier_annot = f\"{brier_med:.2f} ({brier_low:.2f} - {brier_up:.2f})\""
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "## Assess model calibration"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "Let's see how well calibrated are the predicted probabilities"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 52,
+   "metadata": {},
+   "outputs": [
+    {
+     "data": {
+      "image/png": "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",
+      "text/plain": [
+       "<Figure size 1400x600 with 2 Axes>"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    }
+   ],
+   "source": [
+    "import matplotlib.pyplot as plt\n",
+    "from modelsight.calibration import hosmer_lemeshow_plot\n",
+    "\n",
+    "# pool validation ground-truths and predicted probabilities in a long vector\n",
+    "gt_val_conc = np.concatenate(cv_results[\"gt_val\"])\n",
+    "probas_val_conc = np.concatenate(cv_results[\"probas_val\"])\n",
+    "\n",
+    "f, ax = plt.subplots(1, 1, figsize=(14,6))\n",
+    "\n",
+    "f, ax = hosmer_lemeshow_plot(gt_val_conc,\n",
+    "                             probas_val_conc,\n",
+    "                             n_bins=10,\n",
+    "                             colors=('#005CAB', '#FFC325'),\n",
+    "                             annotate_bars=True,\n",
+    "                             title=\"\",\n",
+    "                             brier_score_annot=brier_annot,\n",
+    "                             ax=ax\n",
+    "                             )\n",
+    "\n",
+    "plt.show()"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "The model appears to be perfectly calibrated, as predicted probabilites match observed ones.\n",
+    "\n",
+    "Median brier score also is very low, indicating a good, calibrated classifier."
+   ]
+  }
+ ],
+ "metadata": {
+  "kernelspec": {
+   "display_name": "Python 3 (ipykernel)",
+   "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": "3.10.11"
+  }
+ },
+ "nbformat": 4,
+ "nbformat_minor": 4
+}