notebooks/cross_validation_sol_01.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "id": "b7702f83",
   "metadata": {},
   "source": [
    "# 📃 Solution for Exercise M2.01\n",
    "\n",
    "The aim of this exercise is to make the following experiments:\n",
    "\n",
    "* train and test a support vector machine classifier through\n",
    "  cross-validation;\n",
    "* study the effect of the parameter gamma of this classifier using a\n",
    "  validation curve;\n",
    "* study if it would be useful in term of classification if we could add new\n",
    "  samples in the dataset using a learning curve.\n",
    "\n",
    "To make these experiments we will first load the blood transfusion dataset."
   ]
  },
  {
   "cell_type": "markdown",
   "id": "39dff54b",
   "metadata": {},
   "source": [
    "<div class=\"admonition note alert alert-info\">\n",
    "<p class=\"first admonition-title\" style=\"font-weight: bold;\">Note</p>\n",
    "<p class=\"last\">If you want a deeper overview regarding this dataset, you can refer to the\n",
    "Appendix - Datasets description section at the end of this MOOC.</p>\n",
    "</div>"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "id": "360b4166",
   "metadata": {
    "ExecuteTime": {
     "end_time": "2021-05-27T07:58:14.730768Z",
     "start_time": "2021-05-27T07:58:14.548258Z"
    }
   },
   "outputs": [],
   "source": [
    "import pandas as pd\n",
    "\n",
    "blood_transfusion = pd.read_csv(\"../datasets/blood_transfusion.csv\")\n",
    "data = blood_transfusion.drop(columns=\"Class\")\n",
    "target = blood_transfusion[\"Class\"]"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "222eaeb8",
   "metadata": {},
   "source": [
    "We will use a support vector machine classifier (SVM). In its most simple\n",
    "form, a SVM classifier is a linear classifier behaving similarly to a\n",
    "logistic regression. Indeed, the optimization used to find the optimal\n",
    "weights of the linear model are different but we don't need to know these\n",
    "details for the exercise.\n",
    "\n",
    "Also, this classifier can become more flexible/expressive by using a\n",
    "so-called kernel making the model becomes non-linear. Again, no requirement\n",
    "regarding the mathematics is required to accomplish this exercise.\n",
    "\n",
    "We will use an RBF kernel where a parameter `gamma` allows to tune the\n",
    "flexibility of the model.\n",
    "\n",
    "First let's create a predictive pipeline made of:\n",
    "\n",
    "* a [`sklearn.preprocessing.StandardScaler`](https://scikit-learn.org/stable/modules/generated/sklearn.preprocessing.StandardScaler.html)\n",
    "  with default parameter;\n",
    "* a [`sklearn.svm.SVC`](https://scikit-learn.org/stable/modules/generated/sklearn.svm.SVC.html)\n",
    "  where the parameter `kernel` could be set to `\"rbf\"`. Note that this is the\n",
    "  default."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "id": "491506be",
   "metadata": {
    "ExecuteTime": {
     "end_time": "2021-05-27T07:58:17.201670Z",
     "start_time": "2021-05-27T07:58:16.967757Z"
    }
   },
   "outputs": [],
   "source": [
    "from sklearn.pipeline import make_pipeline\n",
    "from sklearn.preprocessing import StandardScaler\n",
    "from sklearn.svm import SVC\n",
    "\n",
    "model = make_pipeline(StandardScaler(), SVC())"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "1fa38fc0",
   "metadata": {},
   "source": [
    "Evaluate the statistical performance of your model by cross-validation with a\n",
    "`ShuffleSplit` scheme. Thus, you can use\n",
    "[`sklearn.model_selection.cross_validate`](https://scikit-learn.org/stable/modules/generated/sklearn.model_selection.cross_validate.html)\n",
    "and pass a [`sklearn.model_selection.ShuffleSplit`](https://scikit-learn.org/stable/modules/generated/sklearn.model_selection.ShuffleSplit.html)\n",
    "to the `cv` parameter. Only fix the `random_state=0` in the `ShuffleSplit`\n",
    "and let the other parameters to the default."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "id": "9fd70a97",
   "metadata": {
    "ExecuteTime": {
     "end_time": "2021-05-27T07:58:19.155858Z",
     "start_time": "2021-05-27T07:58:18.249510Z"
    }
   },
   "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>fit_time</th>\n",
       "      <th>score_time</th>\n",
       "      <th>test_score</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>0.026244</td>\n",
       "      <td>0.004364</td>\n",
       "      <td>0.680000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>0.025240</td>\n",
       "      <td>0.004335</td>\n",
       "      <td>0.746667</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>0.024766</td>\n",
       "      <td>0.004510</td>\n",
       "      <td>0.786667</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>0.024554</td>\n",
       "      <td>0.004352</td>\n",
       "      <td>0.800000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>0.025450</td>\n",
       "      <td>0.004470</td>\n",
       "      <td>0.746667</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5</th>\n",
       "      <td>0.024876</td>\n",
       "      <td>0.004681</td>\n",
       "      <td>0.786667</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>6</th>\n",
       "      <td>0.025218</td>\n",
       "      <td>0.004508</td>\n",
       "      <td>0.800000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>7</th>\n",
       "      <td>0.023727</td>\n",
       "      <td>0.004345</td>\n",
       "      <td>0.826667</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>8</th>\n",
       "      <td>0.019896</td>\n",
       "      <td>0.002422</td>\n",
       "      <td>0.746667</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>9</th>\n",
       "      <td>0.015910</td>\n",
       "      <td>0.002668</td>\n",
       "      <td>0.733333</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   fit_time  score_time  test_score\n",
       "0  0.026244    0.004364    0.680000\n",
       "1  0.025240    0.004335    0.746667\n",
       "2  0.024766    0.004510    0.786667\n",
       "3  0.024554    0.004352    0.800000\n",
       "4  0.025450    0.004470    0.746667\n",
       "5  0.024876    0.004681    0.786667\n",
       "6  0.025218    0.004508    0.800000\n",
       "7  0.023727    0.004345    0.826667\n",
       "8  0.019896    0.002422    0.746667\n",
       "9  0.015910    0.002668    0.733333"
      ]
     },
     "execution_count": 3,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "from sklearn.model_selection import cross_validate, ShuffleSplit\n",
    "\n",
    "cv = ShuffleSplit(random_state=0)\n",
    "cv_results = cross_validate(model, data, target, cv=cv, n_jobs=-1)\n",
    "cv_results = pd.DataFrame(cv_results)\n",
    "cv_results"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "id": "9cc23559",
   "metadata": {
    "ExecuteTime": {
     "end_time": "2021-05-27T07:58:22.553369Z",
     "start_time": "2021-05-27T07:58:22.550229Z"
    }
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Accuracy score of our model:\n",
      "0.765 +/- 0.043\n"
     ]
    }
   ],
   "source": [
    "print(\n",
    "    f\"Accuracy score of our model:\\n\"\n",
    "    f\"{cv_results['test_score'].mean():.3f} +/- \"\n",
    "    f\"{cv_results['test_score'].std():.3f}\"\n",
    ")"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "ca8432a3",
   "metadata": {},
   "source": [
    "As previously mentioned, the parameter `gamma` is one of the parameter\n",
    "controlling under/over-fitting in support vector machine with an RBF kernel.\n",
    "\n",
    "Compute the validation curve to evaluate the effect of the parameter `gamma`.\n",
    "You can vary its value between `10e-3` and `10e2` by generating samples on a\n",
    "logarithmic scale. Thus, you can use `np.logspace(-3, 2, num=30)`.\n",
    "\n",
    "Since we are manipulating a `Pipeline` the parameter name will be set to\n",
    "`svc__gamma` instead of only `gamma`. You can retrieve the parameter name\n",
    "using `model.get_params().keys()`. We will go more into details regarding\n",
    "accessing and setting hyperparameter in the next section."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "id": "4e14f9a7",
   "metadata": {
    "ExecuteTime": {
     "end_time": "2021-05-27T07:58:38.148711Z",
     "start_time": "2021-05-27T07:58:36.475803Z"
    }
   },
   "outputs": [],
   "source": [
    "import numpy as np\n",
    "from sklearn.model_selection import validation_curve\n",
    "\n",
    "gammas = np.logspace(-3, 2, num=30)\n",
    "param_name = \"svc__gamma\"\n",
    "train_scores, test_scores = validation_curve(\n",
    "    model, data, target, param_name=param_name, param_range=gammas, cv=cv,\n",
    "    n_jobs=-1)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "72bebfc6",
   "metadata": {},
   "source": [
    "Plot the validation curve for the train and test scores."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "id": "b9b3d5c8",
   "metadata": {
    "ExecuteTime": {
     "end_time": "2021-05-27T07:58:40.418855Z",
     "start_time": "2021-05-27T07:58:40.273227Z"
    }
   },
   "outputs": [],
   "source": [
    "import matplotlib.pyplot as plt"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "id": "589d68b4",
   "metadata": {
    "ExecuteTime": {
     "end_time": "2021-05-27T07:58:41.609771Z",
     "start_time": "2021-05-27T07:58:41.391729Z"
    }
   },
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.errorbar(gammas, train_scores.mean(axis=1),\n",
    "             yerr=train_scores.std(axis=1), label='Training error')\n",
    "plt.errorbar(gammas, test_scores.mean(axis=1),\n",
    "             yerr=test_scores.std(axis=1), label='Testing error')\n",
    "plt.legend()\n",
    "\n",
    "plt.xscale(\"log\")\n",
    "plt.xlabel(r\"Value of hyperparameter $\\gamma$\")\n",
    "plt.ylabel(\"Accuracy score\")\n",
    "_ = plt.title(\"Validation score of support vector machine\")"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "7737641e",
   "metadata": {},
   "source": [
    "Looking at the curve, we can clearly identify the over-fitting regime of\n",
    "the SVC classifier when `gamma > 1`.\n",
    "The best setting is around `gamma = 1` while for `gamma < 1`,\n",
    "it is not very clear if the classifier is under-fitting but the\n",
    "testing score is worse than for `gamma = 1`."
   ]
  },
  {
   "cell_type": "markdown",
   "id": "58be261d",
   "metadata": {},
   "source": [
    "Now, you can perform an analysis to check whether adding new samples to the\n",
    "dataset could help our model to better generalize. Compute the learning curve\n",
    "(using [`sklearn.model_selection.learning_curve`](https://scikit-learn.org/stable/modules/generated/sklearn.model_selection.learning_curve.html))\n",
    "by computing the train and test scores for different training dataset size.\n",
    "Plot the train and test scores with respect to the number of samples."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "id": "3e2da762",
   "metadata": {
    "ExecuteTime": {
     "end_time": "2021-05-27T08:03:04.058741Z",
     "start_time": "2021-05-27T08:03:03.755762Z"
    }
   },
   "outputs": [],
   "source": [
    "from sklearn.model_selection import learning_curve\n",
    "\n",
    "train_sizes = np.linspace(0.1, 1, num=10)\n",
    "results = learning_curve(\n",
    "    model, data, target, train_sizes=train_sizes, cv=cv, n_jobs=-1)\n",
    "train_size, train_scores, test_scores = results[:3]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "id": "753962b8",
   "metadata": {
    "ExecuteTime": {
     "end_time": "2021-05-27T08:03:06.183423Z",
     "start_time": "2021-05-27T08:03:06.099870Z"
    }
   },
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.errorbar(train_size, train_scores.mean(axis=1),\n",
    "             yerr=train_scores.std(axis=1), label='Training error')\n",
    "plt.errorbar(train_size, test_scores.mean(axis=1),\n",
    "             yerr=test_scores.std(axis=1), label='Testing error')\n",
    "plt.legend()\n",
    "\n",
    "plt.xlabel(\"Number of samples in the training set\")\n",
    "plt.ylabel(\"Accuracy\")\n",
    "_ = plt.title(\"Learning curve for support vector machine\")"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "9de74847",
   "metadata": {},
   "source": [
    "We observe that adding new samples in the dataset does not improve the\n",
    "testing score. We can only conclude that the standard deviation of\n",
    "the training error is decreasing when adding more samples which is not a\n",
    "surprise."
   ]
  }
 ],
 "metadata": {
  "jupytext": {
   "cell_metadata_filter": "-all",
   "main_language": "python",
   "notebook_metadata_filter": "-all"
  },
  "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": "3.9.4"
  },
  "toc": {
   "base_numbering": 1,
   "nav_menu": {},
   "number_sections": true,
   "sideBar": true,
   "skip_h1_title": false,
   "title_cell": "Table of Contents",
   "title_sidebar": "Contents",
   "toc_cell": false,
   "toc_position": {},
   "toc_section_display": true,
   "toc_window_display": true
  }
 },
 "nbformat": 4,
 "nbformat_minor": 5
}