{"cells": [{"cell_type": "markdown", "metadata": {}, "source": ["# Jeu de donn\u00e9es avec des cat\u00e9gories\n", "\n", "Le jeu de donn\u00e9es [Adult Data Set](https://archive.ics.uci.edu/ml/datasets/adult) ne contient presque que des cat\u00e9gories. Ce notebook explore diff\u00e9rentes moyens de les traiter."]}, {"cell_type": "code", "execution_count": 1, "metadata": {}, "outputs": [], "source": ["%matplotlib inline"]}, {"cell_type": "code", "execution_count": 2, "metadata": {}, "outputs": [{"data": {"text/html": ["<div id=\"my_id_menu_nb\">run previous cell, wait for 2 seconds</div>\n", "<script>\n", "function repeat_indent_string(n){\n", "    var a = \"\" ;\n", "    for ( ; n > 0 ; --n)\n", "        a += \"    \";\n", "    return a;\n", "}\n", "// look up into all sections and builds an automated menu //\n", "var update_menu_string = function(begin, lfirst, llast, sformat, send, keep_item, begin_format, end_format) {\n", "    var anchors = document.getElementsByClassName(\"section\");\n", "    if (anchors.length == 0) {\n", "        anchors = document.getElementsByClassName(\"text_cell_render rendered_html\");\n", "    }\n", "    var i,t;\n", "    var text_menu = begin;\n", "    var text_memo = \"<pre>\\nlength:\" + anchors.length + \"\\n\";\n", "    var ind = \"\";\n", "    var memo_level = 1;\n", "    var href;\n", "    var tags = [];\n", "    var main_item = 0;\n", "    var format_open = 0;\n", "    for (i = 0; i <= llast; i++)\n", "        tags.push(\"h\" + i);\n", "\n", "    for (i = 0; i < anchors.length; i++) {\n", "        text_memo += \"**\" + anchors[i].id + \"--\\n\";\n", "\n", "        var child = null;\n", "        for(t = 0; t < tags.length; t++) {\n", "            var r = anchors[i].getElementsByTagName(tags[t]);\n", "            if (r.length > 0) {\n", "child = r[0];\n", "break;\n", "            }\n", "        }\n", "        if (child == null) {\n", "            text_memo += \"null\\n\";\n", "            continue;\n", "        }\n", "        if (anchors[i].hasAttribute(\"id\")) {\n", "            // when converted in RST\n", "            href = anchors[i].id;\n", "            text_memo += \"#1-\" + href;\n", "            // passer \u00e0 child suivant (le chercher)\n", "        }\n", "        else if (child.hasAttribute(\"id\")) {\n", "            // in a notebook\n", "            href = child.id;\n", "            text_memo += \"#2-\" + href;\n", "        }\n", "        else {\n", "            text_memo += \"#3-\" + \"*\" + \"\\n\";\n", "            continue;\n", "        }\n", "        var title = child.textContent;\n", "        var level = parseInt(child.tagName.substring(1,2));\n", "\n", "        text_memo += \"--\" + level + \"?\" + lfirst + \"--\" + title + \"\\n\";\n", "\n", "        if ((level < lfirst) || (level > llast)) {\n", "            continue ;\n", "        }\n", "        if (title.endsWith('\u00b6')) {\n", "            title = title.substring(0,title.length-1).replace(\"<\", \"&lt;\")\n", "         .replace(\">\", \"&gt;\").replace(\"&\", \"&amp;\");\n", "        }\n", "        if (title.length == 0) {\n", "            continue;\n", "        }\n", "\n", "        while (level < memo_level) {\n", "            text_menu += end_format + \"</ul>\\n\";\n", "            format_open -= 1;\n", "            memo_level -= 1;\n", "        }\n", "        if (level == lfirst) {\n", "            main_item += 1;\n", "        }\n", "        if (keep_item != -1 && main_item != keep_item + 1) {\n", "            // alert(main_item + \" - \" + level + \" - \" + keep_item);\n", "            continue;\n", "        }\n", "        while (level > memo_level) {\n", "            text_menu += \"<ul>\\n\";\n", "            memo_level += 1;\n", "        }\n", "        text_menu += repeat_indent_string(level-2);\n", "        text_menu += begin_format + sformat.replace(\"__HREF__\", href).replace(\"__TITLE__\", title);\n", "        format_open += 1;\n", "    }\n", "    while (1 < memo_level) {\n", "        text_menu += end_format + \"</ul>\\n\";\n", "        memo_level -= 1;\n", "        format_open -= 1;\n", "    }\n", "    text_menu += send;\n", "    //text_menu += \"\\n\" + text_memo;\n", "\n", "    while (format_open > 0) {\n", "        text_menu += end_format;\n", "        format_open -= 1;\n", "    }\n", "    return text_menu;\n", "};\n", "var update_menu = function() {\n", "    var sbegin = \"\";\n", "    var sformat = '<a href=\"#__HREF__\">__TITLE__</a>';\n", "    var send = \"\";\n", "    var begin_format = '<li>';\n", "    var end_format = '</li>';\n", "    var keep_item = -1;\n", "    var text_menu = update_menu_string(sbegin, 2, 4, sformat, send, keep_item,\n", "       begin_format, end_format);\n", "    var menu = document.getElementById(\"my_id_menu_nb\");\n", "    menu.innerHTML=text_menu;\n", "};\n", "window.setTimeout(update_menu,2000);\n", "            </script>"], "text/plain": ["<IPython.core.display.HTML object>"]}, "execution_count": 3, "metadata": {}, "output_type": "execute_result"}], "source": ["from jyquickhelper import add_notebook_menu\n", "add_notebook_menu()"]}, {"cell_type": "code", "execution_count": 3, "metadata": {}, "outputs": [], "source": ["import fiona"]}, {"cell_type": "markdown", "metadata": {}, "source": ["## donn\u00e9es"]}, {"cell_type": "code", "execution_count": 4, "metadata": {"scrolled": false}, "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>age</th>\n", "      <th>workclass</th>\n", "      <th>fnlwgt</th>\n", "      <th>education</th>\n", "      <th>education_num</th>\n", "      <th>marital_status</th>\n", "      <th>occupation</th>\n", "      <th>relationship</th>\n", "      <th>race</th>\n", "      <th>sex</th>\n", "      <th>capital_gain</th>\n", "      <th>capital_loss</th>\n", "      <th>hours_per_week</th>\n", "      <th>native_country</th>\n", "      <th>&lt;=50K</th>\n", "    </tr>\n", "  </thead>\n", "  <tbody>\n", "    <tr>\n", "      <th>0</th>\n", "      <td>39</td>\n", "      <td>State-gov</td>\n", "      <td>77516</td>\n", "      <td>Bachelors</td>\n", "      <td>13</td>\n", "      <td>Never-married</td>\n", "      <td>Adm-clerical</td>\n", "      <td>Not-in-family</td>\n", "      <td>White</td>\n", "      <td>Male</td>\n", "      <td>2174</td>\n", "      <td>0</td>\n", "      <td>40</td>\n", "      <td>United-States</td>\n", "      <td>&lt;=50K</td>\n", "    </tr>\n", "    <tr>\n", "      <th>1</th>\n", "      <td>50</td>\n", "      <td>Self-emp-not-inc</td>\n", "      <td>83311</td>\n", "      <td>Bachelors</td>\n", "      <td>13</td>\n", "      <td>Married-civ-spouse</td>\n", "      <td>Exec-managerial</td>\n", "      <td>Husband</td>\n", "      <td>White</td>\n", "      <td>Male</td>\n", "      <td>0</td>\n", "      <td>0</td>\n", "      <td>13</td>\n", "      <td>United-States</td>\n", "      <td>&lt;=50K</td>\n", "    </tr>\n", "    <tr>\n", "      <th>2</th>\n", "      <td>38</td>\n", "      <td>Private</td>\n", "      <td>215646</td>\n", "      <td>HS-grad</td>\n", "      <td>9</td>\n", "      <td>Divorced</td>\n", "      <td>Handlers-cleaners</td>\n", "      <td>Not-in-family</td>\n", "      <td>White</td>\n", "      <td>Male</td>\n", "      <td>0</td>\n", "      <td>0</td>\n", "      <td>40</td>\n", "      <td>United-States</td>\n", "      <td>&lt;=50K</td>\n", "    </tr>\n", "    <tr>\n", "      <th>3</th>\n", "      <td>53</td>\n", "      <td>Private</td>\n", "      <td>234721</td>\n", "      <td>11th</td>\n", "      <td>7</td>\n", "      <td>Married-civ-spouse</td>\n", "      <td>Handlers-cleaners</td>\n", "      <td>Husband</td>\n", "      <td>Black</td>\n", "      <td>Male</td>\n", "      <td>0</td>\n", "      <td>0</td>\n", "      <td>40</td>\n", "      <td>United-States</td>\n", "      <td>&lt;=50K</td>\n", "    </tr>\n", "    <tr>\n", "      <th>4</th>\n", "      <td>28</td>\n", "      <td>Private</td>\n", "      <td>338409</td>\n", "      <td>Bachelors</td>\n", "      <td>13</td>\n", "      <td>Married-civ-spouse</td>\n", "      <td>Prof-specialty</td>\n", "      <td>Wife</td>\n", "      <td>Black</td>\n", "      <td>Female</td>\n", "      <td>0</td>\n", "      <td>0</td>\n", "      <td>40</td>\n", "      <td>Cuba</td>\n", "      <td>&lt;=50K</td>\n", "    </tr>\n", "  </tbody>\n", "</table>\n", "</div>"], "text/plain": ["   age         workclass  fnlwgt  education  education_num  \\\n", "0   39         State-gov   77516  Bachelors             13   \n", "1   50  Self-emp-not-inc   83311  Bachelors             13   \n", "2   38           Private  215646    HS-grad              9   \n", "3   53           Private  234721       11th              7   \n", "4   28           Private  338409  Bachelors             13   \n", "\n", "       marital_status         occupation   relationship   race     sex  \\\n", "0       Never-married       Adm-clerical  Not-in-family  White    Male   \n", "1  Married-civ-spouse    Exec-managerial        Husband  White    Male   \n", "2            Divorced  Handlers-cleaners  Not-in-family  White    Male   \n", "3  Married-civ-spouse  Handlers-cleaners        Husband  Black    Male   \n", "4  Married-civ-spouse     Prof-specialty           Wife  Black  Female   \n", "\n", "   capital_gain  capital_loss  hours_per_week native_country  <=50K  \n", "0          2174             0              40  United-States  <=50K  \n", "1             0             0              13  United-States  <=50K  \n", "2             0             0              40  United-States  <=50K  \n", "3             0             0              40  United-States  <=50K  \n", "4             0             0              40           Cuba  <=50K  "]}, "execution_count": 5, "metadata": {}, "output_type": "execute_result"}], "source": ["from papierstat.datasets import load_adult_dataset\n", "train, test = load_adult_dataset(url=\"copy\")\n", "train.head()"]}, {"cell_type": "code", "execution_count": 5, "metadata": {}, "outputs": [{"data": {"text/plain": ["{'<=50K', '>50K'}"]}, "execution_count": 6, "metadata": {}, "output_type": "execute_result"}], "source": ["label = '<=50K'\n", "set(train[label])"]}, {"cell_type": "code", "execution_count": 6, "metadata": {}, "outputs": [{"data": {"text/plain": ["{'<=50K', '>50K'}"]}, "execution_count": 7, "metadata": {}, "output_type": "execute_result"}], "source": ["set(test[label])"]}, {"cell_type": "code", "execution_count": 7, "metadata": {}, "outputs": [], "source": ["import numpy\n", "import pandas\n", "X_train = train.drop(label, axis=1)\n", "y_train = train[label] == '>50K'\n", "y_train = pandas.Series(numpy.array([1.0 if y else 0.0 for y in y_train]))\n", "X_test = test.drop(label, axis=1)\n", "y_test = test[label] == '>50K'\n", "y_test = pandas.Series(numpy.array([1.0 if y else 0.0 for y in y_test]))"]}, {"cell_type": "code", "execution_count": 8, "metadata": {}, "outputs": [{"data": {"text/plain": ["age                int64\n", "workclass         object\n", "fnlwgt             int64\n", "education         object\n", "education_num      int64\n", "marital_status    object\n", "occupation        object\n", "relationship      object\n", "race              object\n", "sex               object\n", "capital_gain       int64\n", "capital_loss       int64\n", "hours_per_week     int64\n", "native_country    object\n", "<=50K             object\n", "dtype: object"]}, "execution_count": 9, "metadata": {}, "output_type": "execute_result"}], "source": ["train.dtypes"]}, {"cell_type": "markdown", "metadata": {}, "source": ["La variable *fnlwgt* repr\u00e9sente une forme de pond\u00e9ration : le nombre d'individus que chaque observation repr\u00e9sente. Elle ne doit pas servir \u00e0 la pr\u00e9diction, comme pond\u00e9ration qu'on ignorera : cette pond\u00e9ration n'est pas li\u00e9e aux donn\u00e9es mais \u00e0 l'\u00e9chantillon et elle est impossible \u00e0 construire. Il faut s'en passer."]}, {"cell_type": "code", "execution_count": 9, "metadata": {}, "outputs": [], "source": ["X_train = X_train.drop(['fnlwgt'], axis=1).copy()\n", "X_test = X_test.drop(['fnlwgt'], axis=1).copy()"]}, {"cell_type": "markdown", "metadata": {}, "source": ["## cat\u00e9gories\n", "\n", "On garde la liste des variables cat\u00e9gorielles."]}, {"cell_type": "code", "execution_count": 10, "metadata": {}, "outputs": [{"data": {"text/plain": ["['workclass',\n", " 'education',\n", " 'marital_status',\n", " 'occupation',\n", " 'relationship',\n", " 'race',\n", " 'sex',\n", " 'native_country']"]}, "execution_count": 11, "metadata": {}, "output_type": "execute_result"}], "source": ["cat_col = list(_ for _ in X_train.select_dtypes(\"object\").columns)\n", "cat_col"]}, {"cell_type": "markdown", "metadata": {}, "source": ["La fonction [get_dummies](https://pandas.pydata.org/pandas-docs/stable/generated/pandas.get_dummies.html) est pratique mais probl\u00e9matique si les modalit\u00e9s de la base d'apprentissage et la base de test sont diff\u00e9rentes, ce qui est fr\u00e9quemment le cas pour les cat\u00e9gories peu fr\u00e9quentes. Il faudrait regrouper les deux bases, l'appliquer puis s\u00e9pareer \u00e0 nouveau. Trop long. On veut utiliser [OneHotEncoder](http://scikit-learn.org/stable/modules/generated/sklearn.preprocessing.OneHotEncoder.html) et [LabelEncoder](http://scikit-learn.org/stable/modules/generated/sklearn.preprocessing.LabelEncoder.html) mais ce n'est pas tr\u00e8s pratique parce que [OneHotEncoder](http://scikit-learn.org/stable/modules/generated/sklearn.preprocessing.OneHotEncoder.html) n'accepte que les colonnes enti\u00e8res et [LabelEncoder](http://scikit-learn.org/stable/modules/generated/sklearn.preprocessing.LabelEncoder.html) et peut traiter qu'une colonne \u00e0 la fois."]}, {"cell_type": "code", "execution_count": 11, "metadata": {}, "outputs": [{"name": "stdout", "output_type": "stream", "text": ["fit_transform() takes 2 positional arguments but 3 were given\n"]}], "source": ["from sklearn.pipeline import make_pipeline\n", "from sklearn.preprocessing import LabelEncoder, OneHotEncoder\n", "pipe = make_pipeline(LabelEncoder(), OneHotEncoder())\n", "try:\n", "    pipe.fit(X_train[cat_col[0]], y_train)\n", "except Exception as e:\n", "    print(e)"]}, {"cell_type": "markdown", "metadata": {}, "source": ["On utilise [OneHotEncoder](http://contrib.scikit-learn.org/categorical-encoding/) du module [category_encoders](http://contrib.scikit-learn.org/categorical-encoding/)."]}, {"cell_type": "code", "execution_count": 12, "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>age</th>\n", "      <th>workclass_1</th>\n", "      <th>workclass_2</th>\n", "      <th>workclass_3</th>\n", "      <th>workclass_4</th>\n", "      <th>workclass_5</th>\n", "      <th>workclass_6</th>\n", "      <th>workclass_7</th>\n", "      <th>workclass_8</th>\n", "      <th>workclass_9</th>\n", "      <th>...</th>\n", "      <th>native_country_33</th>\n", "      <th>native_country_34</th>\n", "      <th>native_country_35</th>\n", "      <th>native_country_36</th>\n", "      <th>native_country_37</th>\n", "      <th>native_country_38</th>\n", "      <th>native_country_39</th>\n", "      <th>native_country_40</th>\n", "      <th>native_country_41</th>\n", "      <th>native_country_42</th>\n", "    </tr>\n", "  </thead>\n", "  <tbody>\n", "    <tr>\n", "      <th>0</th>\n", "      <td>39</td>\n", "      <td>1</td>\n", "      <td>0</td>\n", "      <td>0</td>\n", "      <td>0</td>\n", "      <td>0</td>\n", "      <td>0</td>\n", "      <td>0</td>\n", "      <td>0</td>\n", "      <td>0</td>\n", "      <td>...</td>\n", "      <td>0</td>\n", "      <td>0</td>\n", "      <td>0</td>\n", "      <td>0</td>\n", "      <td>0</td>\n", "      <td>0</td>\n", "      <td>0</td>\n", "      <td>0</td>\n", "      <td>0</td>\n", "      <td>0</td>\n", "    </tr>\n", "    <tr>\n", "      <th>1</th>\n", "      <td>50</td>\n", "      <td>0</td>\n", "      <td>1</td>\n", "      <td>0</td>\n", "      <td>0</td>\n", "      <td>0</td>\n", "      <td>0</td>\n", "      <td>0</td>\n", "      <td>0</td>\n", "      <td>0</td>\n", "      <td>...</td>\n", "      <td>0</td>\n", "      <td>0</td>\n", "      <td>0</td>\n", "      <td>0</td>\n", "      <td>0</td>\n", "      <td>0</td>\n", "      <td>0</td>\n", "      <td>0</td>\n", "      <td>0</td>\n", "      <td>0</td>\n", "    </tr>\n", "    <tr>\n", "      <th>2</th>\n", "      <td>38</td>\n", "      <td>0</td>\n", "      <td>0</td>\n", "      <td>1</td>\n", "      <td>0</td>\n", "      <td>0</td>\n", "      <td>0</td>\n", "      <td>0</td>\n", "      <td>0</td>\n", "      <td>0</td>\n", "      <td>...</td>\n", "      <td>0</td>\n", "      <td>0</td>\n", "      <td>0</td>\n", "      <td>0</td>\n", "      <td>0</td>\n", "      <td>0</td>\n", "      <td>0</td>\n", "      <td>0</td>\n", "      <td>0</td>\n", "      <td>0</td>\n", "    </tr>\n", "    <tr>\n", "      <th>3</th>\n", "      <td>53</td>\n", "      <td>0</td>\n", "      <td>0</td>\n", "      <td>1</td>\n", "      <td>0</td>\n", "      <td>0</td>\n", "      <td>0</td>\n", "      <td>0</td>\n", "      <td>0</td>\n", "      <td>0</td>\n", "      <td>...</td>\n", "      <td>0</td>\n", "      <td>0</td>\n", "      <td>0</td>\n", "      <td>0</td>\n", "      <td>0</td>\n", "      <td>0</td>\n", "      <td>0</td>\n", "      <td>0</td>\n", "      <td>0</td>\n", "      <td>0</td>\n", "    </tr>\n", "    <tr>\n", "      <th>4</th>\n", "      <td>28</td>\n", "      <td>0</td>\n", "      <td>0</td>\n", "      <td>1</td>\n", "      <td>0</td>\n", "      <td>0</td>\n", "      <td>0</td>\n", "      <td>0</td>\n", "      <td>0</td>\n", "      <td>0</td>\n", "      <td>...</td>\n", "      <td>0</td>\n", "      <td>0</td>\n", "      <td>0</td>\n", "      <td>0</td>\n", "      <td>0</td>\n", "      <td>0</td>\n", "      <td>0</td>\n", "      <td>0</td>\n", "      <td>0</td>\n", "      <td>0</td>\n", "    </tr>\n", "  </tbody>\n", "</table>\n", "<p>5 rows \u00d7 107 columns</p>\n", "</div>"], "text/plain": ["   age  workclass_1  workclass_2  workclass_3  workclass_4  workclass_5  \\\n", "0   39            1            0            0            0            0   \n", "1   50            0            1            0            0            0   \n", "2   38            0            0            1            0            0   \n", "3   53            0            0            1            0            0   \n", "4   28            0            0            1            0            0   \n", "\n", "   workclass_6  workclass_7  workclass_8  workclass_9  ...  native_country_33  \\\n", "0            0            0            0            0  ...                  0   \n", "1            0            0            0            0  ...                  0   \n", "2            0            0            0            0  ...                  0   \n", "3            0            0            0            0  ...                  0   \n", "4            0            0            0            0  ...                  0   \n", "\n", "   native_country_34  native_country_35  native_country_36  native_country_37  \\\n", "0                  0                  0                  0                  0   \n", "1                  0                  0                  0                  0   \n", "2                  0                  0                  0                  0   \n", "3                  0                  0                  0                  0   \n", "4                  0                  0                  0                  0   \n", "\n", "   native_country_38  native_country_39  native_country_40  native_country_41  \\\n", "0                  0                  0                  0                  0   \n", "1                  0                  0                  0                  0   \n", "2                  0                  0                  0                  0   \n", "3                  0                  0                  0                  0   \n", "4                  0                  0                  0                  0   \n", "\n", "   native_country_42  \n", "0                  0  \n", "1                  0  \n", "2                  0  \n", "3                  0  \n", "4                  0  \n", "\n", "[5 rows x 107 columns]"]}, "execution_count": 13, "metadata": {}, "output_type": "execute_result"}], "source": ["from category_encoders import OneHotEncoder\n", "ce = OneHotEncoder(cols=cat_col, handle_missing='value',\n", "                   drop_invariant=False, handle_unknown='value')\n", "X_train_cat = ce.fit_transform(X_train)\n", "X_train_cat.head()"]}, {"cell_type": "markdown", "metadata": {}, "source": ["C'est assez compliqu\u00e9 \u00e0 lire. On renomme les colonnes avec les noms des cat\u00e9gories originaux. Il y a probablement mieux mais ce code fonctionne."]}, {"cell_type": "code", "execution_count": 13, "metadata": {"scrolled": false}, "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>age</th>\n", "      <th>workclass__Self-emp-not-inc</th>\n", "      <th>workclass__Private</th>\n", "      <th>workclass__Federal-gov</th>\n", "      <th>workclass__Local-gov</th>\n", "      <th>workclass__?</th>\n", "      <th>workclass__Self-emp-inc</th>\n", "      <th>workclass__Without-pay</th>\n", "      <th>workclass__Never-worked</th>\n", "      <th>workclass__nan</th>\n", "      <th>...</th>\n", "      <th>native_country__Scotland</th>\n", "      <th>native_country__Trinadad&amp;Tobago</th>\n", "      <th>native_country__Greece</th>\n", "      <th>native_country__Nicaragua</th>\n", "      <th>native_country__Vietnam</th>\n", "      <th>native_country__Hong</th>\n", "      <th>native_country__Ireland</th>\n", "      <th>native_country__Hungary</th>\n", "      <th>native_country__Holand-Netherlands</th>\n", "      <th>native_country__nan</th>\n", "    </tr>\n", "  </thead>\n", "  <tbody>\n", "    <tr>\n", "      <th>0</th>\n", "      <td>39</td>\n", "      <td>1</td>\n", "      <td>0</td>\n", "      <td>0</td>\n", "      <td>0</td>\n", "      <td>0</td>\n", "      <td>0</td>\n", "      <td>0</td>\n", "      <td>0</td>\n", "      <td>0</td>\n", "      <td>...</td>\n", "      <td>0</td>\n", "      <td>0</td>\n", "      <td>0</td>\n", "      <td>0</td>\n", "      <td>0</td>\n", "      <td>0</td>\n", "      <td>0</td>\n", "      <td>0</td>\n", "      <td>0</td>\n", "      <td>0</td>\n", "    </tr>\n", "    <tr>\n", "      <th>1</th>\n", "      <td>50</td>\n", "      <td>0</td>\n", "      <td>1</td>\n", "      <td>0</td>\n", "      <td>0</td>\n", "      <td>0</td>\n", "      <td>0</td>\n", "      <td>0</td>\n", "      <td>0</td>\n", "      <td>0</td>\n", "      <td>...</td>\n", "      <td>0</td>\n", "      <td>0</td>\n", "      <td>0</td>\n", "      <td>0</td>\n", "      <td>0</td>\n", "      <td>0</td>\n", "      <td>0</td>\n", "      <td>0</td>\n", "      <td>0</td>\n", "      <td>0</td>\n", "    </tr>\n", "    <tr>\n", "      <th>2</th>\n", "      <td>38</td>\n", "      <td>0</td>\n", "      <td>0</td>\n", "      <td>1</td>\n", "      <td>0</td>\n", "      <td>0</td>\n", "      <td>0</td>\n", "      <td>0</td>\n", "      <td>0</td>\n", "      <td>0</td>\n", "      <td>...</td>\n", "      <td>0</td>\n", "      <td>0</td>\n", "      <td>0</td>\n", "      <td>0</td>\n", "      <td>0</td>\n", "      <td>0</td>\n", "      <td>0</td>\n", "      <td>0</td>\n", "      <td>0</td>\n", "      <td>0</td>\n", "    </tr>\n", "    <tr>\n", "      <th>3</th>\n", "      <td>53</td>\n", "      <td>0</td>\n", "      <td>0</td>\n", "      <td>1</td>\n", "      <td>0</td>\n", "      <td>0</td>\n", "      <td>0</td>\n", "      <td>0</td>\n", "      <td>0</td>\n", "      <td>0</td>\n", "      <td>...</td>\n", "      <td>0</td>\n", "      <td>0</td>\n", "      <td>0</td>\n", "      <td>0</td>\n", "      <td>0</td>\n", "      <td>0</td>\n", "      <td>0</td>\n", "      <td>0</td>\n", "      <td>0</td>\n", "      <td>0</td>\n", "    </tr>\n", "    <tr>\n", "      <th>4</th>\n", "      <td>28</td>\n", "      <td>0</td>\n", "      <td>0</td>\n", "      <td>1</td>\n", "      <td>0</td>\n", "      <td>0</td>\n", "      <td>0</td>\n", "      <td>0</td>\n", "      <td>0</td>\n", "      <td>0</td>\n", "      <td>...</td>\n", "      <td>0</td>\n", "      <td>0</td>\n", "      <td>0</td>\n", "      <td>0</td>\n", "      <td>0</td>\n", "      <td>0</td>\n", "      <td>0</td>\n", "      <td>0</td>\n", "      <td>0</td>\n", "      <td>0</td>\n", "    </tr>\n", "  </tbody>\n", "</table>\n", "<p>5 rows \u00d7 107 columns</p>\n", "</div>"], "text/plain": ["   age  workclass__Self-emp-not-inc  workclass__Private  \\\n", "0   39                            1                   0   \n", "1   50                            0                   1   \n", "2   38                            0                   0   \n", "3   53                            0                   0   \n", "4   28                            0                   0   \n", "\n", "   workclass__Federal-gov  workclass__Local-gov  workclass__?  \\\n", "0                       0                     0             0   \n", "1                       0                     0             0   \n", "2                       1                     0             0   \n", "3                       1                     0             0   \n", "4                       1                     0             0   \n", "\n", "   workclass__Self-emp-inc  workclass__Without-pay  workclass__Never-worked  \\\n", "0                        0                       0                        0   \n", "1                        0                       0                        0   \n", "2                        0                       0                        0   \n", "3                        0                       0                        0   \n", "4                        0                       0                        0   \n", "\n", "   workclass__nan  ...  native_country__Scotland  \\\n", "0               0  ...                         0   \n", "1               0  ...                         0   \n", "2               0  ...                         0   \n", "3               0  ...                         0   \n", "4               0  ...                         0   \n", "\n", "   native_country__Trinadad&Tobago  native_country__Greece  \\\n", "0                                0                       0   \n", "1                                0                       0   \n", "2                                0                       0   \n", "3                                0                       0   \n", "4                                0                       0   \n", "\n", "   native_country__Nicaragua  native_country__Vietnam  native_country__Hong  \\\n", "0                          0                        0                     0   \n", "1                          0                        0                     0   \n", "2                          0                        0                     0   \n", "3                          0                        0                     0   \n", "4                          0                        0                     0   \n", "\n", "   native_country__Ireland  native_country__Hungary  \\\n", "0                        0                        0   \n", "1                        0                        0   \n", "2                        0                        0   \n", "3                        0                        0   \n", "4                        0                        0   \n", "\n", "   native_country__Holand-Netherlands  native_country__nan  \n", "0                                   0                    0  \n", "1                                   0                    0  \n", "2                                   0                    0  \n", "3                                   0                    0  \n", "4                                   0                    0  \n", "\n", "[5 rows x 107 columns]"]}, "execution_count": 14, "metadata": {}, "output_type": "execute_result"}], "source": ["def rename_columns(df, ce):\n", "    rev_mapping = {r['col']: r['mapping'] for r in ce.category_mapping}\n", "    cols = []\n", "    for c in df.columns:\n", "        if '_' not in c:\n", "            cols.append(c)\n", "            continue\n", "        spl = c.split('_')\n", "        col = \"_\".join(spl[:-1])\n", "        try:\n", "            nb = int(spl[-1])    \n", "            mapping  = rev_mapping[col]        \n", "            cols.append(str(col) + \"__\" + str(mapping.index[nb]))\n", "        except ValueError:\n", "            cols.append(c)\n", "    df.columns = cols + list(df.columns)[len(cols):]\n", "\n", "rename_columns(X_train_cat, ce)\n", "X_train_cat.head()"]}, {"cell_type": "markdown", "metadata": {}, "source": ["C'est plus clair. Une derni\u00e8re remarque sur le param\u00e8tre *handle_missing* et *handle_unknown*. Le premier d\u00e9finit la fa\u00e7on de g\u00e9rer les valeurs manquantes, le premier la fa\u00e7on dont le mod\u00e8le doit g\u00e9rer les cat\u00e9gories nouvelles, c'est-\u00e0-dire les cat\u00e9gories non vues dans la base d'apprentissage."]}, {"cell_type": "markdown", "metadata": {}, "source": ["## premier jet\n", "\n", "On construit un [pipeline](http://scikit-learn.org/stable/modules/generated/sklearn.pipeline.Pipeline.html)."]}, {"cell_type": "code", "execution_count": 14, "metadata": {}, "outputs": [{"name": "stderr", "output_type": "stream", "text": ["C:\\xavierdupre\\__home_\\github_fork\\scikit-learn\\sklearn\\linear_model\\_logistic.py:764: ConvergenceWarning: lbfgs failed to converge (status=1):\n", "STOP: TOTAL NO. of ITERATIONS REACHED LIMIT.\n", "\n", "Increase the number of iterations (max_iter) or scale the data as shown in:\n", "    https://scikit-learn.org/stable/modules/preprocessing.html\n", "Please also refer to the documentation for alternative solver options:\n", "    https://scikit-learn.org/stable/modules/linear_model.html#logistic-regression\n", "  extra_warning_msg=_LOGISTIC_SOLVER_CONVERGENCE_MSG)\n"]}, {"data": {"text/plain": ["Pipeline(memory=None,\n", "         steps=[('onehotencoder',\n", "                 OneHotEncoder(cols=['workclass', 'education', 'marital_status',\n", "                                     'occupation', 'relationship', 'race',\n", "                                     'sex', 'native_country'],\n", "                               drop_invariant=False, handle_missing='value',\n", "                               handle_unknown='value', return_df=True,\n", "                               use_cat_names=False, verbose=0)),\n", "                ('logisticregression',\n", "                 LogisticRegression(C=1.0, class_weight=None, dual=False,\n", "                                    fit_intercept=True, intercept_scaling=1,\n", "                                    l1_ratio=None, max_iter=100,\n", "                                    multi_class='auto', n_jobs=None,\n", "                                    penalty='l2', random_state=None,\n", "                                    solver='lbfgs', tol=0.0001, verbose=0,\n", "                                    warm_start=False))],\n", "         verbose=False)"]}, "execution_count": 15, "metadata": {}, "output_type": "execute_result"}], "source": ["from sklearn.linear_model import LogisticRegression\n", "pipe = make_pipeline(\n", "            OneHotEncoder(cols=cat_col, handle_missing='value',\n", "                          drop_invariant=False, handle_unknown='value'), \n", "            LogisticRegression())\n", "pipe.fit(X_train, y_train)"]}, {"cell_type": "code", "execution_count": 15, "metadata": {}, "outputs": [{"data": {"text/plain": ["0.8426386585590566"]}, "execution_count": 16, "metadata": {}, "output_type": "execute_result"}], "source": ["pipe.score(X_test, y_test)"]}, {"cell_type": "markdown", "metadata": {}, "source": ["On essaye avec une [RandomForest](http://scikit-learn.org/stable/modules/generated/sklearn.ensemble.RandomForestClassifier.html)."]}, {"cell_type": "code", "execution_count": 16, "metadata": {}, "outputs": [{"data": {"text/plain": ["Pipeline(memory=None,\n", "         steps=[('onehotencoder',\n", "                 OneHotEncoder(cols=['workclass', 'education', 'marital_status',\n", "                                     'occupation', 'relationship', 'race',\n", "                                     'sex', 'native_country'],\n", "                               drop_invariant=False, handle_missing='value',\n", "                               handle_unknown='value', return_df=True,\n", "                               use_cat_names=False, verbose=0)),\n", "                ('randomforestclassifier',\n", "                 RandomForestClassifier(bootstrap=True, ccp_alpha=0.0,\n", "                                        class_weight=None, criterion='gini',\n", "                                        max_depth=None, max_features='auto',\n", "                                        max_leaf_nodes=None, max_samples=None,\n", "                                        min_impurity_decrease=0.0,\n", "                                        min_impurity_split=None,\n", "                                        min_samples_leaf=1, min_samples_split=2,\n", "                                        min_weight_fraction_leaf=0.0,\n", "                                        n_estimators=100, n_jobs=None,\n", "                                        oob_score=False, random_state=None,\n", "                                        verbose=0, warm_start=False))],\n", "         verbose=False)"]}, "execution_count": 17, "metadata": {}, "output_type": "execute_result"}], "source": ["from sklearn.ensemble import RandomForestClassifier\n", "pipe2 = make_pipeline(ce, RandomForestClassifier(n_estimators=100))\n", "pipe2.fit(X_train, y_train)"]}, {"cell_type": "code", "execution_count": 17, "metadata": {}, "outputs": [{"data": {"text/plain": ["0.8448498249493275"]}, "execution_count": 18, "metadata": {}, "output_type": "execute_result"}], "source": ["pipe2.score(X_test, y_test)"]}, {"cell_type": "code", "execution_count": 18, "metadata": {}, "outputs": [{"data": {"text/plain": ["array([0.22779679, 0.00529596, 0.00965327, 0.01187745, 0.00584904])"]}, "execution_count": 19, "metadata": {}, "output_type": "execute_result"}], "source": ["pipe2.steps[-1][-1].feature_importances_[:5]"]}, {"cell_type": "markdown", "metadata": {}, "source": ["On regarde l'[importance des features](http://scikit-learn.org/stable/auto_examples/ensemble/plot_forest_importances.html)."]}, {"cell_type": "code", "execution_count": 19, "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>importance</th>\n", "    </tr>\n", "    <tr>\n", "      <th>name</th>\n", "      <th></th>\n", "    </tr>\n", "  </thead>\n", "  <tbody>\n", "    <tr>\n", "      <th>age</th>\n", "      <td>0.227797</td>\n", "    </tr>\n", "    <tr>\n", "      <th>hours_per_week</th>\n", "      <td>0.113923</td>\n", "    </tr>\n", "    <tr>\n", "      <th>capital_gain</th>\n", "      <td>0.110967</td>\n", "    </tr>\n", "    <tr>\n", "      <th>marital_status__Divorced</th>\n", "      <td>0.067712</td>\n", "    </tr>\n", "    <tr>\n", "      <th>education_num</th>\n", "      <td>0.064527</td>\n", "    </tr>\n", "  </tbody>\n", "</table>\n", "</div>"], "text/plain": ["                          importance\n", "name                                \n", "age                         0.227797\n", "hours_per_week              0.113923\n", "capital_gain                0.110967\n", "marital_status__Divorced    0.067712\n", "education_num               0.064527"]}, "execution_count": 20, "metadata": {}, "output_type": "execute_result"}], "source": ["import pandas\n", "df = pandas.DataFrame(dict(name=X_train_cat.columns, \n", "                           importance=pipe2.steps[-1][-1].feature_importances_))\n", "df = df.sort_values(\"importance\", ascending=False).reset_index(drop=True)\n", "df = df.set_index('name')\n", "df.head()"]}, {"cell_type": "code", "execution_count": 20, "metadata": {"scrolled": false}, "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": ["import matplotlib.pyplot as plt\n", "fig, ax = plt.subplots(1, 1, figsize=(6, 4))\n", "df[:10].plot.barh(ax=ax)\n", "ax.set_title('Importance des variables - RandomForest');"]}, {"cell_type": "markdown", "metadata": {}, "source": ["On compare avec [XGBoost](http://xgboost.readthedocs.io/en/latest/model.html). L'\u00e2ge semble la variable plus importante. Cela dit, si ce graphique donne quelques pistes, ce n'est pas la v\u00e9rit\u00e9 car les variables peuvent \u00eatre corr\u00e9l\u00e9es. Deux variables corr\u00e9l\u00e9es sont interchangeables."]}, {"cell_type": "code", "execution_count": 21, "metadata": {}, "outputs": [{"data": {"text/plain": ["Pipeline(memory=None,\n", "         steps=[('onehotencoder',\n", "                 OneHotEncoder(cols=['workclass', 'education', 'marital_status',\n", "                                     'occupation', 'relationship', 'race',\n", "                                     'sex', 'native_country'],\n", "                               drop_invariant=False, handle_missing='value',\n", "                               handle_unknown='value', return_df=True,\n", "                               use_cat_names=False, verbose=0)),\n", "                ('xgbclassifier',\n", "                 XGBClassifier(base_score=0.5, booster='gbtree',\n", "                               colsample_bylevel=1, colsample_bynode=1,\n", "                               colsample_bytree=1, gamma=0, learning_rate=0.1,\n", "                               max_delta_step=0, max_depth=3,\n", "                               min_child_weight=1, missing=None,\n", "                               n_estimators=100, n_jobs=1, nthread=None,\n", "                               objective='binary:logistic', random_state=0,\n", "                               reg_alpha=0, reg_lambda=1, scale_pos_weight=1,\n", "                               seed=None, silent=None, subsample=1,\n", "                               verbosity=1))],\n", "         verbose=False)"]}, "execution_count": 22, "metadata": {}, "output_type": "execute_result"}], "source": ["from xgboost import XGBClassifier\n", "pipe3 = make_pipeline(ce, XGBClassifier())\n", "pipe3.fit(X_train, y_train)"]}, {"cell_type": "code", "execution_count": 22, "metadata": {}, "outputs": [{"data": {"text/plain": ["0.869172655242307"]}, "execution_count": 23, "metadata": {}, "output_type": "execute_result"}], "source": ["pipe3.score(X_test, y_test)"]}, {"cell_type": "code", "execution_count": 23, "metadata": {}, "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": ["df = pandas.DataFrame(dict(name=X_train_cat.columns, \n", "                           importance=pipe3.steps[-1][-1].feature_importances_))\n", "df = df.sort_values(\"importance\", ascending=False).reset_index(drop=True)\n", "df = df.set_index('name')\n", "fig, ax = plt.subplots(1, 1, figsize=(6, 4))\n", "df[:10].plot.barh(ax=ax)\n", "ax.set_title('Importance des variables - XGBoost');"]}, {"cell_type": "markdown", "metadata": {}, "source": ["On retrouve presque les m\u00eames variables mais pas dans le m\u00eame ordre. On essaye un dernier module [catboost](https://tech.yandex.com/catboost/)."]}, {"cell_type": "code", "execution_count": 24, "metadata": {}, "outputs": [{"data": {"text/plain": ["Pipeline(memory=None,\n", "         steps=[('onehotencoder',\n", "                 OneHotEncoder(cols=['workclass', 'education', 'marital_status',\n", "                                     'occupation', 'relationship', 'race',\n", "                                     'sex', 'native_country'],\n", "                               drop_invariant=False, handle_missing='value',\n", "                               handle_unknown='value', return_df=True,\n", "                               use_cat_names=False, verbose=0)),\n", "                ('catboostclassifier',\n", "                 <catboost.core.CatBoostClassifier object at 0x000001E0DF767AC8>)],\n", "         verbose=False)"]}, "execution_count": 25, "metadata": {}, "output_type": "execute_result"}], "source": ["from catboost import CatBoostClassifier\n", "pipe4 = make_pipeline(ce, CatBoostClassifier(iterations=100, verbose=False))\n", "pipe4.fit(X_train, y_train)"]}, {"cell_type": "code", "execution_count": 25, "metadata": {}, "outputs": [{"data": {"text/plain": ["0.0"]}, "execution_count": 26, "metadata": {}, "output_type": "execute_result"}], "source": ["pipe4.score(X_test, y_test)"]}, {"cell_type": "code", "execution_count": 26, "metadata": {}, "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": ["df = pandas.DataFrame(dict(name=X_train_cat.columns, \n", "                           importance=pipe4.steps[-1][-1].feature_importances_))\n", "df = df.sort_values(\"importance\", ascending=False).reset_index(drop=True)\n", "df = df.set_index('name')\n", "fig, ax = plt.subplots(1, 1, figsize=(6, 4))\n", "df[:10].plot.barh(ax=ax)\n", "ax.set_title('Importance des variables - CatBoost');"]}, {"cell_type": "markdown", "metadata": {}, "source": ["Les mod\u00e8les sont \u00e0 peu pr\u00e8s d'accord sur la performance mais pas vraiment sur l'ordre des features les plus importantes. Comme ce sont tous des random forests, m\u00eame apprises diff\u00e9remment, on peut supposer qu'il existe des corr\u00e9lations entre les variables. Des corr\u00e9lations au sens du mod\u00e8le donc pas n\u00e9cessairement lin\u00e9aires."]}, {"cell_type": "markdown", "metadata": {}, "source": ["## Courbe ROC"]}, {"cell_type": "code", "execution_count": 27, "metadata": {}, "outputs": [], "source": ["from sklearn.metrics import roc_curve\n", "import warnings\n", "if len(pipe2.steps[-1][-1].classes_) != len(pipe3.steps[-1][-1].classes_):\n", "    raise Exception(\"M\u00e9sentente classificatoire pipe2 {0} != pipe3 {1}\".format(\n", "                    pipe2.steps[-1][-1].classes_, pipe3.steps[-1][-1].classes_))\n", "if len(pipe2.steps[-1][-1].classes_) != len(pipe4.steps[-1][-1].classes_):\n", "    if not pipe4.steps[-1][-1].classes_:\n", "        # Probably a bug (happens on circleci).\n", "        # Assuming classes are in the same order.\n", "        # See https://github.com/catboost/catboost/blob/master/catboost/python-package/catboost/core.py#L1994\n", "        warnings.warn(\"pipe4.steps[-1][-1].classes_ is empty.\")\n", "        pipe4.steps[-1][-1]._classes = pipe2.steps[-1][-1].classes_\n", "if len(pipe2.steps[-1][-1].classes_) != len(pipe4.steps[-1][-1].classes_):\n", "    print(\"M\u00e9sentente classificatoire pipe2 {0} != pipe4 {1}\".format(\n", "                    pipe2.steps[-1][-1].classes_, pipe4.steps[-1][-1].classes_))\n", "\n", "index2 = pipe2.steps[-1][-1].classes_[1]\n", "index3 = pipe3.steps[-1][-1].classes_[1]\n", "index4 = pipe4.steps[-1][-1].classes_[1]\n", "fpr2, tpr2, th2 = roc_curve(y_test, pipe2.predict_proba(X_test)[:, 1], \n", "                            pos_label=index2, drop_intermediate=False)\n", "fpr3, tpr3, th3 = roc_curve(y_test, pipe3.predict_proba(X_test)[:, 1], \n", "                            pos_label=index3, drop_intermediate=False)\n", "if len(pipe4.steps[-1][-1].classes_) >= 2:\n", "    fpr4, tpr4, th4 = roc_curve(y_test, pipe4.predict_proba(X_test)[:, 1], \n", "                                pos_label=index4, drop_intermediate=False)\n", "else:\n", "    fpr4 = None"]}, {"cell_type": "code", "execution_count": 28, "metadata": {}, "outputs": [{"data": {"image/png": "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\n", "text/plain": ["<Figure size 288x288 with 1 Axes>"]}, "metadata": {"needs_background": "light"}, "output_type": "display_data"}], "source": ["from sklearn.metrics import auc\n", "fig, ax = plt.subplots(1, 1, figsize=(4, 4))\n", "ax.plot(fpr2, tpr2, label='%1.3f RandomForest' % auc(fpr2, tpr2), color='y')\n", "ax.plot(fpr3, tpr3, label='%1.3f XGBoost' % auc(fpr3, tpr3))\n", "if fpr4 is not None:\n", "    ax.plot(fpr4, tpr4, label='%1.3f CatBoost' % auc(fpr4, tpr4))\n", "ax.legend()\n", "ax.set_title('Courbe ROC pour trois mod\u00e8les');"]}, {"cell_type": "markdown", "metadata": {}, "source": ["## GridSearch\n", "\n", "On cherche \u00e0 optimiser les hyperparam\u00e8tres sur la base d'apprentissage. On v\u00e9rifie d'abord que les donn\u00e9es sont bien identiquement distribu\u00e9es. La validation crois\u00e9e consid\u00e8re des parties contig\u00fces de la base de donn\u00e9es. Il arrive que les donn\u00e9es ne soient pas tout-\u00e0-fait homog\u00e8nes et qu'il faille les m\u00e9langer. On compare les performances avant et apr\u00e8s m\u00e9lange pour v\u00e9rifier que l'ordre des donn\u00e9es n'a pas d'incidence."]}, {"cell_type": "code", "execution_count": 29, "metadata": {"scrolled": false}, "outputs": [{"data": {"text/plain": ["array([0.84707508, 0.84152334, 0.84259828, 0.85288698, 0.84766585])"]}, "execution_count": 30, "metadata": {}, "output_type": "execute_result"}], "source": ["from sklearn.model_selection import cross_val_score\n", "cross_val_score(pipe2, X_train, y_train, cv=5)"]}, {"cell_type": "code", "execution_count": 30, "metadata": {"scrolled": false}, "outputs": [{"data": {"text/plain": ["array([0.84277599, 0.84797297, 0.84659091, 0.84981572, 0.85089066])"]}, "execution_count": 31, "metadata": {}, "output_type": "execute_result"}], "source": ["from pandas_streaming.df import dataframe_shuffle\n", "from numpy.random import permutation\n", "index = permutation(X_train.index)\n", "X_train_shuffled = X_train.iloc[index, :]\n", "y_train_shuffled = y_train[index]\n", "cross_val_score(pipe2, X_train_shuffled, y_train_shuffled, cv=5)"]}, {"cell_type": "markdown", "metadata": {}, "source": ["L'ordre de la base n'a pas d'incidence."]}, {"cell_type": "code", "execution_count": 31, "metadata": {"scrolled": false}, "outputs": [{"name": "stdout", "output_type": "stream", "text": ["Fitting 5 folds for each of 6 candidates, totalling 30 fits\n", "[CV] randomforestclassifier__min_samples_leaf=2, randomforestclassifier__n_estimators=10 \n"]}, {"name": "stderr", "output_type": "stream", "text": ["[Parallel(n_jobs=1)]: Using backend SequentialBackend with 1 concurrent workers.\n"]}, {"name": "stdout", "output_type": "stream", "text": ["[CV]  randomforestclassifier__min_samples_leaf=2, randomforestclassifier__n_estimators=10, total=   1.2s\n", "[CV] randomforestclassifier__min_samples_leaf=2, randomforestclassifier__n_estimators=10 \n"]}, {"name": "stderr", "output_type": "stream", "text": ["[Parallel(n_jobs=1)]: Done   1 out of   1 | elapsed:    1.1s remaining:    0.0s\n"]}, {"name": "stdout", "output_type": "stream", "text": ["[CV]  randomforestclassifier__min_samples_leaf=2, randomforestclassifier__n_estimators=10, total=   1.2s\n", "[CV] randomforestclassifier__min_samples_leaf=2, randomforestclassifier__n_estimators=10 \n", "[CV]  randomforestclassifier__min_samples_leaf=2, randomforestclassifier__n_estimators=10, total=   1.2s\n", "[CV] randomforestclassifier__min_samples_leaf=2, randomforestclassifier__n_estimators=10 \n", "[CV]  randomforestclassifier__min_samples_leaf=2, randomforestclassifier__n_estimators=10, total=   1.2s\n", "[CV] randomforestclassifier__min_samples_leaf=2, randomforestclassifier__n_estimators=10 \n", "[CV]  randomforestclassifier__min_samples_leaf=2, randomforestclassifier__n_estimators=10, total=   1.2s\n", "[CV] randomforestclassifier__min_samples_leaf=2, randomforestclassifier__n_estimators=20 \n", "[CV]  randomforestclassifier__min_samples_leaf=2, randomforestclassifier__n_estimators=20, total=   1.6s\n", "[CV] randomforestclassifier__min_samples_leaf=2, randomforestclassifier__n_estimators=20 \n", "[CV]  randomforestclassifier__min_samples_leaf=2, randomforestclassifier__n_estimators=20, total=   1.6s\n", "[CV] randomforestclassifier__min_samples_leaf=2, randomforestclassifier__n_estimators=20 \n", "[CV]  randomforestclassifier__min_samples_leaf=2, randomforestclassifier__n_estimators=20, total=   1.6s\n", "[CV] randomforestclassifier__min_samples_leaf=2, randomforestclassifier__n_estimators=20 \n", "[CV]  randomforestclassifier__min_samples_leaf=2, randomforestclassifier__n_estimators=20, total=   1.5s\n", "[CV] randomforestclassifier__min_samples_leaf=2, randomforestclassifier__n_estimators=20 \n", "[CV]  randomforestclassifier__min_samples_leaf=2, randomforestclassifier__n_estimators=20, total=   1.6s\n", "[CV] randomforestclassifier__min_samples_leaf=2, randomforestclassifier__n_estimators=50 \n", "[CV]  randomforestclassifier__min_samples_leaf=2, randomforestclassifier__n_estimators=50, total=   2.6s\n", "[CV] randomforestclassifier__min_samples_leaf=2, randomforestclassifier__n_estimators=50 \n", "[CV]  randomforestclassifier__min_samples_leaf=2, randomforestclassifier__n_estimators=50, total=   3.2s\n", "[CV] randomforestclassifier__min_samples_leaf=2, randomforestclassifier__n_estimators=50 \n", "[CV]  randomforestclassifier__min_samples_leaf=2, randomforestclassifier__n_estimators=50, total=   3.4s\n", "[CV] randomforestclassifier__min_samples_leaf=2, randomforestclassifier__n_estimators=50 \n", "[CV]  randomforestclassifier__min_samples_leaf=2, randomforestclassifier__n_estimators=50, total=   3.6s\n", "[CV] randomforestclassifier__min_samples_leaf=2, randomforestclassifier__n_estimators=50 \n", "[CV]  randomforestclassifier__min_samples_leaf=2, randomforestclassifier__n_estimators=50, total=   2.8s\n", "[CV] randomforestclassifier__min_samples_leaf=10, randomforestclassifier__n_estimators=10 \n", "[CV]  randomforestclassifier__min_samples_leaf=10, randomforestclassifier__n_estimators=10, total=   1.4s\n", "[CV] randomforestclassifier__min_samples_leaf=10, randomforestclassifier__n_estimators=10 \n", "[CV]  randomforestclassifier__min_samples_leaf=10, randomforestclassifier__n_estimators=10, total=   1.2s\n", "[CV] randomforestclassifier__min_samples_leaf=10, randomforestclassifier__n_estimators=10 \n", "[CV]  randomforestclassifier__min_samples_leaf=10, randomforestclassifier__n_estimators=10, total=   1.1s\n", "[CV] randomforestclassifier__min_samples_leaf=10, randomforestclassifier__n_estimators=10 \n", "[CV]  randomforestclassifier__min_samples_leaf=10, randomforestclassifier__n_estimators=10, total=   1.1s\n", "[CV] randomforestclassifier__min_samples_leaf=10, randomforestclassifier__n_estimators=10 \n", "[CV]  randomforestclassifier__min_samples_leaf=10, randomforestclassifier__n_estimators=10, total=   1.1s\n", "[CV] randomforestclassifier__min_samples_leaf=10, randomforestclassifier__n_estimators=20 \n", "[CV]  randomforestclassifier__min_samples_leaf=10, randomforestclassifier__n_estimators=20, total=   1.4s\n", "[CV] randomforestclassifier__min_samples_leaf=10, randomforestclassifier__n_estimators=20 \n", "[CV]  randomforestclassifier__min_samples_leaf=10, randomforestclassifier__n_estimators=20, total=   1.4s\n", "[CV] randomforestclassifier__min_samples_leaf=10, randomforestclassifier__n_estimators=20 \n", "[CV]  randomforestclassifier__min_samples_leaf=10, randomforestclassifier__n_estimators=20, total=   1.5s\n", "[CV] randomforestclassifier__min_samples_leaf=10, randomforestclassifier__n_estimators=20 \n", "[CV]  randomforestclassifier__min_samples_leaf=10, randomforestclassifier__n_estimators=20, total=   1.6s\n", "[CV] randomforestclassifier__min_samples_leaf=10, randomforestclassifier__n_estimators=20 \n", "[CV]  randomforestclassifier__min_samples_leaf=10, randomforestclassifier__n_estimators=20, total=   1.6s\n", "[CV] randomforestclassifier__min_samples_leaf=10, randomforestclassifier__n_estimators=50 \n", "[CV]  randomforestclassifier__min_samples_leaf=10, randomforestclassifier__n_estimators=50, total=   2.4s\n", "[CV] randomforestclassifier__min_samples_leaf=10, randomforestclassifier__n_estimators=50 \n", "[CV]  randomforestclassifier__min_samples_leaf=10, randomforestclassifier__n_estimators=50, total=   2.8s\n", "[CV] randomforestclassifier__min_samples_leaf=10, randomforestclassifier__n_estimators=50 \n", "[CV]  randomforestclassifier__min_samples_leaf=10, randomforestclassifier__n_estimators=50, total=   2.9s\n", "[CV] randomforestclassifier__min_samples_leaf=10, randomforestclassifier__n_estimators=50 \n", "[CV]  randomforestclassifier__min_samples_leaf=10, randomforestclassifier__n_estimators=50, total=   3.9s\n", "[CV] randomforestclassifier__min_samples_leaf=10, randomforestclassifier__n_estimators=50 \n", "[CV]  randomforestclassifier__min_samples_leaf=10, randomforestclassifier__n_estimators=50, total=   3.1s\n"]}, {"name": "stderr", "output_type": "stream", "text": ["[Parallel(n_jobs=1)]: Done  30 out of  30 | elapsed:   58.3s finished\n"]}, {"data": {"text/plain": ["GridSearchCV(cv=None, error_score=nan,\n", "             estimator=Pipeline(memory=None,\n", "                                steps=[('onehotencoder',\n", "                                        OneHotEncoder(cols=['workclass',\n", "                                                            'education',\n", "                                                            'marital_status',\n", "                                                            'occupation',\n", "                                                            'relationship',\n", "                                                            'race', 'sex',\n", "                                                            'native_country'],\n", "                                                      drop_invariant=False,\n", "                                                      handle_missing='value',\n", "                                                      handle_unknown='value',\n", "                                                      return_df=True,\n", "                                                      use_cat_names=False,\n", "                                                      verbose=0)),\n", "                                       ('randomforestclassifier...\n", "                                                               min_weight_fraction_leaf=0.0,\n", "                                                               n_estimators=100,\n", "                                                               n_jobs=None,\n", "                                                               oob_score=False,\n", "                                                               random_state=None,\n", "                                                               verbose=0,\n", "                                                               warm_start=False))],\n", "                                verbose=False),\n", "             iid='deprecated', n_jobs=None,\n", "             param_grid={'randomforestclassifier__min_samples_leaf': [2, 10],\n", "                         'randomforestclassifier__n_estimators': [10, 20, 50]},\n", "             pre_dispatch='2*n_jobs', refit=True, return_train_score=False,\n", "             scoring=None, verbose=2)"]}, "execution_count": 32, "metadata": {}, "output_type": "execute_result"}], "source": ["from sklearn.model_selection import GridSearchCV\n", "param_grid = {'randomforestclassifier__n_estimators':[10, 20, 50],\n", "              'randomforestclassifier__min_samples_leaf': [2, 10]}\n", "cvgrid = GridSearchCV(estimator=pipe2, param_grid=param_grid, verbose=2)\n", "cvgrid.fit(X_train, y_train)"]}, {"cell_type": "code", "execution_count": 32, "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>randomforestclassifier__min_samples_leaf</th>\n", "      <th>randomforestclassifier__n_estimators</th>\n", "      <th>mean_fit_time</th>\n", "      <th>mean_test_score</th>\n", "    </tr>\n", "  </thead>\n", "  <tbody>\n", "    <tr>\n", "      <th>3</th>\n", "      <td>10</td>\n", "      <td>10</td>\n", "      <td>1.069937</td>\n", "      <td>0.857437</td>\n", "    </tr>\n", "    <tr>\n", "      <th>4</th>\n", "      <td>10</td>\n", "      <td>20</td>\n", "      <td>1.351183</td>\n", "      <td>0.858635</td>\n", "    </tr>\n", "    <tr>\n", "      <th>5</th>\n", "      <td>10</td>\n", "      <td>50</td>\n", "      <td>2.862142</td>\n", "      <td>0.859034</td>\n", "    </tr>\n", "    <tr>\n", "      <th>0</th>\n", "      <td>2</td>\n", "      <td>10</td>\n", "      <td>1.094900</td>\n", "      <td>0.861092</td>\n", "    </tr>\n", "    <tr>\n", "      <th>1</th>\n", "      <td>2</td>\n", "      <td>20</td>\n", "      <td>1.456303</td>\n", "      <td>0.863641</td>\n", "    </tr>\n", "    <tr>\n", "      <th>2</th>\n", "      <td>2</td>\n", "      <td>50</td>\n", "      <td>2.944120</td>\n", "      <td>0.863764</td>\n", "    </tr>\n", "  </tbody>\n", "</table>\n", "</div>"], "text/plain": ["   randomforestclassifier__min_samples_leaf  \\\n", "3                                        10   \n", "4                                        10   \n", "5                                        10   \n", "0                                         2   \n", "1                                         2   \n", "2                                         2   \n", "\n", "   randomforestclassifier__n_estimators  mean_fit_time  mean_test_score  \n", "3                                    10       1.069937         0.857437  \n", "4                                    20       1.351183         0.858635  \n", "5                                    50       2.862142         0.859034  \n", "0                                    10       1.094900         0.861092  \n", "1                                    20       1.456303         0.863641  \n", "2                                    50       2.944120         0.863764  "]}, "execution_count": 33, "metadata": {}, "output_type": "execute_result"}], "source": ["import pandas\n", "df = pandas.DataFrame(cvgrid.cv_results_['params'])\n", "df['mean_fit_time'] = cvgrid.cv_results_['mean_fit_time']\n", "df['mean_test_score'] = cvgrid.cv_results_['mean_test_score']\n", "df.sort_values('mean_test_score')"]}, {"cell_type": "markdown", "metadata": {}, "source": ["Il faudrait continuer \u00e0 explorer les hyperparam\u00e8tres et confirmer sur la base de test. A priori, cela marche mieux avec plus d'arbres."]}, {"cell_type": "markdown", "metadata": {}, "source": ["## Features polyn\u00f4miales\n", "\n", "On essaye m\u00eame si cela a peu de chance d'aboutir compte tenu des variables, principalement cat\u00e9gorielles, et du fait qu'on utilise une for\u00eat al\u00e9atoire."]}, {"cell_type": "code", "execution_count": 33, "metadata": {}, "outputs": [{"data": {"text/plain": ["Pipeline(memory=None,\n", "         steps=[('onehotencoder',\n", "                 OneHotEncoder(cols=['workclass', 'education', 'marital_status',\n", "                                     'occupation', 'relationship', 'race',\n", "                                     'sex', 'native_country'],\n", "                               drop_invariant=False, handle_missing='value',\n", "                               handle_unknown='value', return_df=True,\n", "                               use_cat_names=False, verbose=0)),\n", "                ('polynomialfeatures',\n", "                 PolynomialFeatures(degree=2, include_bias=True,\n", "                                    int...\n", "                 RandomForestClassifier(bootstrap=True, ccp_alpha=0.0,\n", "                                        class_weight=None, criterion='gini',\n", "                                        max_depth=None, max_features='auto',\n", "                                        max_leaf_nodes=None, max_samples=None,\n", "                                        min_impurity_decrease=0.0,\n", "                                        min_impurity_split=None,\n", "                                        min_samples_leaf=1, min_samples_split=2,\n", "                                        min_weight_fraction_leaf=0.0,\n", "                                        n_estimators=100, n_jobs=None,\n", "                                        oob_score=False, random_state=None,\n", "                                        verbose=0, warm_start=False))],\n", "         verbose=False)"]}, "execution_count": 34, "metadata": {}, "output_type": "execute_result"}], "source": ["from sklearn.preprocessing import PolynomialFeatures\n", "pipe5 = make_pipeline(ce, PolynomialFeatures(), RandomForestClassifier())\n", "pipe5.fit(X_train, y_train)"]}, {"cell_type": "code", "execution_count": 34, "metadata": {}, "outputs": [{"data": {"text/plain": ["0.8463853571647934"]}, "execution_count": 35, "metadata": {}, "output_type": "execute_result"}], "source": ["pipe5.score(X_test, y_test)"]}, {"cell_type": "markdown", "metadata": {}, "source": ["Ca n'am\u00e9liore pas."]}, {"cell_type": "markdown", "metadata": {}, "source": ["## Interpr\u00e9tation\n", "\n", "On souhaite en savoir plus sur les variables."]}, {"cell_type": "code", "execution_count": 35, "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>age</th>\n", "      <th>workclass__Self-emp-not-inc</th>\n", "      <th>workclass__Private</th>\n", "      <th>workclass__Federal-gov</th>\n", "      <th>workclass__Local-gov</th>\n", "      <th>workclass__?</th>\n", "      <th>workclass__Self-emp-inc</th>\n", "      <th>workclass__Without-pay</th>\n", "      <th>workclass__Never-worked</th>\n", "      <th>workclass__nan</th>\n", "      <th>...</th>\n", "      <th>native_country__Trinadad&amp;Tobago</th>\n", "      <th>native_country__Greece</th>\n", "      <th>native_country__Nicaragua</th>\n", "      <th>native_country__Vietnam</th>\n", "      <th>native_country__Hong</th>\n", "      <th>native_country__Ireland</th>\n", "      <th>native_country__Hungary</th>\n", "      <th>native_country__Holand-Netherlands</th>\n", "      <th>native_country__nan</th>\n", "      <th>0</th>\n", "    </tr>\n", "  </thead>\n", "  <tbody>\n", "    <tr>\n", "      <th>0</th>\n", "      <td>39</td>\n", "      <td>1</td>\n", "      <td>0</td>\n", "      <td>0</td>\n", "      <td>0</td>\n", "      <td>0</td>\n", "      <td>0</td>\n", "      <td>0</td>\n", "      <td>0</td>\n", "      <td>0</td>\n", "      <td>...</td>\n", "      <td>0</td>\n", "      <td>0</td>\n", "      <td>0</td>\n", "      <td>0</td>\n", "      <td>0</td>\n", "      <td>0</td>\n", "      <td>0</td>\n", "      <td>0</td>\n", "      <td>0</td>\n", "      <td>0.0</td>\n", "    </tr>\n", "    <tr>\n", "      <th>1</th>\n", "      <td>50</td>\n", "      <td>0</td>\n", "      <td>1</td>\n", "      <td>0</td>\n", "      <td>0</td>\n", "      <td>0</td>\n", "      <td>0</td>\n", "      <td>0</td>\n", "      <td>0</td>\n", "      <td>0</td>\n", "      <td>...</td>\n", "      <td>0</td>\n", "      <td>0</td>\n", "      <td>0</td>\n", "      <td>0</td>\n", "      <td>0</td>\n", "      <td>0</td>\n", "      <td>0</td>\n", "      <td>0</td>\n", "      <td>0</td>\n", "      <td>0.0</td>\n", "    </tr>\n", "    <tr>\n", "      <th>2</th>\n", "      <td>38</td>\n", "      <td>0</td>\n", "      <td>0</td>\n", "      <td>1</td>\n", "      <td>0</td>\n", "      <td>0</td>\n", "      <td>0</td>\n", "      <td>0</td>\n", "      <td>0</td>\n", "      <td>0</td>\n", "      <td>...</td>\n", "      <td>0</td>\n", "      <td>0</td>\n", "      <td>0</td>\n", "      <td>0</td>\n", "      <td>0</td>\n", "      <td>0</td>\n", "      <td>0</td>\n", "      <td>0</td>\n", "      <td>0</td>\n", "      <td>0.0</td>\n", "    </tr>\n", "    <tr>\n", "      <th>3</th>\n", "      <td>53</td>\n", "      <td>0</td>\n", "      <td>0</td>\n", "      <td>1</td>\n", "      <td>0</td>\n", "      <td>0</td>\n", "      <td>0</td>\n", "      <td>0</td>\n", "      <td>0</td>\n", "      <td>0</td>\n", "      <td>...</td>\n", "      <td>0</td>\n", "      <td>0</td>\n", "      <td>0</td>\n", "      <td>0</td>\n", "      <td>0</td>\n", "      <td>0</td>\n", "      <td>0</td>\n", "      <td>0</td>\n", "      <td>0</td>\n", "      <td>0.0</td>\n", "    </tr>\n", "    <tr>\n", "      <th>4</th>\n", "      <td>28</td>\n", "      <td>0</td>\n", "      <td>0</td>\n", "      <td>1</td>\n", "      <td>0</td>\n", "      <td>0</td>\n", "      <td>0</td>\n", "      <td>0</td>\n", "      <td>0</td>\n", "      <td>0</td>\n", "      <td>...</td>\n", "      <td>0</td>\n", "      <td>0</td>\n", "      <td>0</td>\n", "      <td>0</td>\n", "      <td>0</td>\n", "      <td>0</td>\n", "      <td>0</td>\n", "      <td>0</td>\n", "      <td>0</td>\n", "      <td>0.0</td>\n", "    </tr>\n", "  </tbody>\n", "</table>\n", "<p>5 rows \u00d7 108 columns</p>\n", "</div>"], "text/plain": ["   age  workclass__Self-emp-not-inc  workclass__Private  \\\n", "0   39                            1                   0   \n", "1   50                            0                   1   \n", "2   38                            0                   0   \n", "3   53                            0                   0   \n", "4   28                            0                   0   \n", "\n", "   workclass__Federal-gov  workclass__Local-gov  workclass__?  \\\n", "0                       0                     0             0   \n", "1                       0                     0             0   \n", "2                       1                     0             0   \n", "3                       1                     0             0   \n", "4                       1                     0             0   \n", "\n", "   workclass__Self-emp-inc  workclass__Without-pay  workclass__Never-worked  \\\n", "0                        0                       0                        0   \n", "1                        0                       0                        0   \n", "2                        0                       0                        0   \n", "3                        0                       0                        0   \n", "4                        0                       0                        0   \n", "\n", "   workclass__nan  ...  native_country__Trinadad&Tobago  \\\n", "0               0  ...                                0   \n", "1               0  ...                                0   \n", "2               0  ...                                0   \n", "3               0  ...                                0   \n", "4               0  ...                                0   \n", "\n", "   native_country__Greece  native_country__Nicaragua  native_country__Vietnam  \\\n", "0                       0                          0                        0   \n", "1                       0                          0                        0   \n", "2                       0                          0                        0   \n", "3                       0                          0                        0   \n", "4                       0                          0                        0   \n", "\n", "   native_country__Hong  native_country__Ireland  native_country__Hungary  \\\n", "0                     0                        0                        0   \n", "1                     0                        0                        0   \n", "2                     0                        0                        0   \n", "3                     0                        0                        0   \n", "4                     0                        0                        0   \n", "\n", "   native_country__Holand-Netherlands  native_country__nan    0  \n", "0                                   0                    0  0.0  \n", "1                                   0                    0  0.0  \n", "2                                   0                    0  0.0  \n", "3                                   0                    0  0.0  \n", "4                                   0                    0  0.0  \n", "\n", "[5 rows x 108 columns]"]}, "execution_count": 36, "metadata": {}, "output_type": "execute_result"}], "source": ["conc = pandas.concat([X_train_cat, pandas.Series(y_train)], axis=1)\n", "conc.head()"]}, {"cell_type": "code", "execution_count": 36, "metadata": {"scrolled": false}, "outputs": [], "source": ["corr = conc.corr()"]}, {"cell_type": "code", "execution_count": 37, "metadata": {}, "outputs": [{"name": "stderr", "output_type": "stream", "text": ["c:\\python372_x64\\lib\\site-packages\\statsmodels\\tools\\_testing.py:19: FutureWarning: pandas.util.testing is deprecated. Use the functions in the public API at pandas.testing instead.\n", "  import pandas.util.testing as tm\n", "c:\\python372_x64\\lib\\site-packages\\seaborn\\matrix.py:624: UserWarning: Clustering large matrix with scipy. Installing `fastcluster` may give better performance.\n", "  warnings.warn(msg)\n"]}, {"data": {"image/png": "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\n", "text/plain": ["<Figure size 1080x1080 with 4 Axes>"]}, "metadata": {"needs_background": "light"}, "output_type": "display_data"}], "source": ["from seaborn import clustermap\n", "\n", "clustermap(corr, center=0, cmap=\"vlag\", linewidths=.75, figsize=(15, 15));"]}, {"cell_type": "markdown", "metadata": {}, "source": ["Ce n'est pas facile \u00e0 voir. Il faudrait essayer avec [bokeh](https://bokeh.pydata.org/en/latest/docs/gallery/les_mis.html) ou essayer de proc\u00e9der autrement."]}, {"cell_type": "markdown", "metadata": {}, "source": ["## ACM\n", "\n", "Ce qui suit n'est pas tout-\u00e0-fait une [ACM](https://fr.wikipedia.org/wiki/Analyse_des_correspondances_multiples) mais cela s'en inspire. On consid\u00e8re les variables comme des observations et on les projette sur des plans d\u00e9finis par les axes d'une [ACP](https://fr.wikipedia.org/wiki/Analyse_en_composantes_principales). On normalise \u00e9galement car on m\u00e9lange variables continues et variables binaires d'ordre de grandeur diff\u00e9rents. Les calculs sont plus pr\u00e9cis lorsque les matrices ont des coefficients de m\u00eame ordre. Le dernier exercice de cet examen [Programmation ENSAE 2006](http://www.xavierdupre.fr/site2013/enseignements/tdnote/ecrit_2006.pdf) ach\u00e8vera de vous convaincre."]}, {"cell_type": "code", "execution_count": 38, "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>0</th>\n", "      <th>1</th>\n", "      <th>2</th>\n", "      <th>3</th>\n", "      <th>4</th>\n", "      <th>5</th>\n", "      <th>6</th>\n", "      <th>7</th>\n", "      <th>8</th>\n", "      <th>9</th>\n", "      <th>...</th>\n", "      <th>32551</th>\n", "      <th>32552</th>\n", "      <th>32553</th>\n", "      <th>32554</th>\n", "      <th>32555</th>\n", "      <th>32556</th>\n", "      <th>32557</th>\n", "      <th>32558</th>\n", "      <th>32559</th>\n", "      <th>32560</th>\n", "    </tr>\n", "  </thead>\n", "  <tbody>\n", "    <tr>\n", "      <th>age</th>\n", "      <td>0.030671</td>\n", "      <td>0.837109</td>\n", "      <td>-0.042642</td>\n", "      <td>1.057047</td>\n", "      <td>-0.775768</td>\n", "      <td>-0.115955</td>\n", "      <td>0.763796</td>\n", "      <td>0.983734</td>\n", "      <td>-0.555830</td>\n", "      <td>0.250608</td>\n", "      <td>...</td>\n", "      <td>-0.482518</td>\n", "      <td>0.323921</td>\n", "      <td>-0.482518</td>\n", "      <td>1.057047</td>\n", "      <td>-1.215643</td>\n", "      <td>-0.849080</td>\n", "      <td>0.103983</td>\n", "      <td>1.423610</td>\n", "      <td>-1.215643</td>\n", "      <td>0.983734</td>\n", "    </tr>\n", "    <tr>\n", "      <th>workclass__Self-emp-not-inc</th>\n", "      <td>4.907700</td>\n", "      <td>-0.203761</td>\n", "      <td>-0.203761</td>\n", "      <td>-0.203761</td>\n", "      <td>-0.203761</td>\n", "      <td>-0.203761</td>\n", "      <td>-0.203761</td>\n", "      <td>-0.203761</td>\n", "      <td>-0.203761</td>\n", "      <td>-0.203761</td>\n", "      <td>...</td>\n", "      <td>-0.203761</td>\n", "      <td>-0.203761</td>\n", "      <td>-0.203761</td>\n", "      <td>-0.203761</td>\n", "      <td>-0.203761</td>\n", "      <td>-0.203761</td>\n", "      <td>-0.203761</td>\n", "      <td>-0.203761</td>\n", "      <td>-0.203761</td>\n", "      <td>-0.203761</td>\n", "    </tr>\n", "  </tbody>\n", "</table>\n", "<p>2 rows \u00d7 32561 columns</p>\n", "</div>"], "text/plain": ["                                0         1         2         3         4      \\\n", "age                          0.030671  0.837109 -0.042642  1.057047 -0.775768   \n", "workclass__Self-emp-not-inc  4.907700 -0.203761 -0.203761 -0.203761 -0.203761   \n", "\n", "                                5         6         7         8         9      \\\n", "age                         -0.115955  0.763796  0.983734 -0.555830  0.250608   \n", "workclass__Self-emp-not-inc -0.203761 -0.203761 -0.203761 -0.203761 -0.203761   \n", "\n", "                             ...     32551     32552     32553     32554  \\\n", "age                          ... -0.482518  0.323921 -0.482518  1.057047   \n", "workclass__Self-emp-not-inc  ... -0.203761 -0.203761 -0.203761 -0.203761   \n", "\n", "                                32555     32556     32557     32558     32559  \\\n", "age                         -1.215643 -0.849080  0.103983  1.423610 -1.215643   \n", "workclass__Self-emp-not-inc -0.203761 -0.203761 -0.203761 -0.203761 -0.203761   \n", "\n", "                                32560  \n", "age                          0.983734  \n", "workclass__Self-emp-not-inc -0.203761  \n", "\n", "[2 rows x 32561 columns]"]}, "execution_count": 39, "metadata": {}, "output_type": "execute_result"}], "source": ["from sklearn.preprocessing import StandardScaler\n", "import pandas\n", "rows_cat = pandas.DataFrame(StandardScaler().fit_transform(X_train_cat))\n", "rows_cat.columns = X_train_cat.columns\n", "rows_cat = rows_cat.T\n", "rows_cat.head(n=2)"]}, {"cell_type": "code", "execution_count": 39, "metadata": {}, "outputs": [{"data": {"text/plain": ["PCA(copy=True, iterated_power='auto', n_components=3, random_state=None,\n", "    svd_solver='auto', tol=0.0, whiten=False)"]}, "execution_count": 40, "metadata": {}, "output_type": "execute_result"}], "source": ["from sklearn.decomposition import PCA\n", "pca = PCA(n_components=3)\n", "pca.fit(rows_cat)"]}, {"cell_type": "code", "execution_count": 40, "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>axe1</th>\n", "      <th>axe2</th>\n", "      <th>axe3</th>\n", "    </tr>\n", "  </thead>\n", "  <tbody>\n", "    <tr>\n", "      <th>sex__nan</th>\n", "      <td>-129.499567</td>\n", "      <td>60.907917</td>\n", "      <td>29.421935</td>\n", "    </tr>\n", "    <tr>\n", "      <th>marital_status__Married-civ-spouse</th>\n", "      <td>-106.135296</td>\n", "      <td>-11.383937</td>\n", "      <td>-54.355587</td>\n", "    </tr>\n", "  </tbody>\n", "</table>\n", "</div>"], "text/plain": ["                                          axe1       axe2       axe3\n", "sex__nan                           -129.499567  60.907917  29.421935\n", "marital_status__Married-civ-spouse -106.135296 -11.383937 -54.355587"]}, "execution_count": 41, "metadata": {}, "output_type": "execute_result"}], "source": ["import pandas\n", "tr = pandas.DataFrame(pca.transform(rows_cat))\n", "tr.columns = ['axe1', 'axe2', 'axe3']\n", "tr.index = rows_cat.index\n", "tr.sort_values('axe1').head(n=2)"]}, {"cell_type": "code", "execution_count": 41, "metadata": {}, "outputs": [{"data": {"image/png": "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\n", "text/plain": ["<Figure size 720x720 with 1 Axes>"]}, "metadata": {"needs_background": "light"}, "output_type": "display_data"}], "source": ["ax = tr.plot(x='axe1', y='axe2', kind='scatter', figsize=(10, 10))\n", "for t, (x, y, z) in tr.iterrows():\n", "    ax.text(x, y, t, fontsize=10, rotation=10)\n", "ax.set_title(\"ACP sur les variables - axe 1, 2\");"]}, {"cell_type": "code", "execution_count": 42, "metadata": {}, "outputs": [{"data": {"image/png": "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\n", "text/plain": ["<Figure size 720x720 with 1 Axes>"]}, "metadata": {"needs_background": "light"}, "output_type": "display_data"}], "source": ["ax = tr.plot(x='axe1', y='axe3', kind='scatter', figsize=(10, 10))\n", "for t, (x, y, z) in tr.iterrows():\n", "    ax.text(x, z, t, fontsize=10, rotation=10)\n", "ax.set_title(\"ACP sur les variables - axe 1, 3\");"]}, {"cell_type": "markdown", "metadata": {}, "source": ["On voit quelques variables \u00e0 supprimer car tr\u00e8s corr\u00e9l\u00e9es comme la relation ou la situation maritale. On voit aussi que les deux genres homme/femme sont oppos\u00e9s. On voit aussi que certaines cat\u00e9gories sont tr\u00e8s proches comme Prof, Masters ou dipl\u00f4m\u00e9s. Il est probable que le mod\u00e8le de pr\u00e9diction ne p\u00e2tisse pas du regroupement de ces trois cat\u00e9gories. On utilise [bokeh](https://bokeh.pydata.org/en/latest/) pour pouvoir zoomer."]}, {"cell_type": "code", "execution_count": 43, "metadata": {}, "outputs": [{"data": {"text/html": ["\n", "    <div class=\"bk-root\">\n", "        <a href=\"https://bokeh.org\" target=\"_blank\" class=\"bk-logo bk-logo-small bk-logo-notebook\"></a>\n", "        <span id=\"1001\">Loading BokehJS ...</span>\n", "    </div>"]}, "metadata": {}, "output_type": "display_data"}, {"data": {"application/javascript": ["\n", "(function(root) {\n", "  function now() {\n", "    return new Date();\n", "  }\n", "\n", "  var force = true;\n", "\n", "  if (typeof root._bokeh_onload_callbacks === \"undefined\" || force === true) {\n", "    root._bokeh_onload_callbacks = [];\n", "    root._bokeh_is_loading = undefined;\n", "  }\n", "\n", "  var JS_MIME_TYPE = 'application/javascript';\n", "  var HTML_MIME_TYPE = 'text/html';\n", "  var EXEC_MIME_TYPE = 'application/vnd.bokehjs_exec.v0+json';\n", "  var CLASS_NAME = 'output_bokeh rendered_html';\n", "\n", "  /**\n", "   * Render data to the DOM node\n", "   */\n", "  function render(props, node) {\n", "    var script = document.createElement(\"script\");\n", "    node.appendChild(script);\n", "  }\n", "\n", "  /**\n", "   * Handle when an output is cleared or removed\n", "   */\n", "  function handleClearOutput(event, handle) {\n", "    var cell = handle.cell;\n", "\n", "    var id = cell.output_area._bokeh_element_id;\n", "    var server_id = cell.output_area._bokeh_server_id;\n", "    // Clean up Bokeh references\n", "    if (id != null && id in Bokeh.index) {\n", "      Bokeh.index[id].model.document.clear();\n", "      delete Bokeh.index[id];\n", "    }\n", "\n", "    if (server_id !== undefined) {\n", "      // Clean up Bokeh references\n", "      var cmd = \"from bokeh.io.state import curstate; print(curstate().uuid_to_server['\" + server_id + \"'].get_sessions()[0].document.roots[0]._id)\";\n", "      cell.notebook.kernel.execute(cmd, {\n", "        iopub: {\n", "          output: function(msg) {\n", "            var id = msg.content.text.trim();\n", "            if (id in Bokeh.index) {\n", "              Bokeh.index[id].model.document.clear();\n", "              delete Bokeh.index[id];\n", "            }\n", "          }\n", "        }\n", "      });\n", "      // Destroy server and session\n", "      var cmd = \"import bokeh.io.notebook as ion; ion.destroy_server('\" + server_id + \"')\";\n", "      cell.notebook.kernel.execute(cmd);\n", "    }\n", "  }\n", "\n", "  /**\n", "   * Handle when a new output is added\n", "   */\n", "  function handleAddOutput(event, handle) {\n", "    var output_area = handle.output_area;\n", "    var output = handle.output;\n", "\n", "    // limit handleAddOutput to display_data with EXEC_MIME_TYPE content only\n", "    if ((output.output_type != \"display_data\") || (!output.data.hasOwnProperty(EXEC_MIME_TYPE))) {\n", "      return\n", "    }\n", "\n", "    var toinsert = output_area.element.find(\".\" + CLASS_NAME.split(' ')[0]);\n", "\n", "    if (output.metadata[EXEC_MIME_TYPE][\"id\"] !== undefined) {\n", "      toinsert[toinsert.length - 1].firstChild.textContent = output.data[JS_MIME_TYPE];\n", "      // store reference to embed id on output_area\n", "      output_area._bokeh_element_id = output.metadata[EXEC_MIME_TYPE][\"id\"];\n", "    }\n", "    if (output.metadata[EXEC_MIME_TYPE][\"server_id\"] !== undefined) {\n", "      var bk_div = document.createElement(\"div\");\n", "      bk_div.innerHTML = output.data[HTML_MIME_TYPE];\n", "      var script_attrs = bk_div.children[0].attributes;\n", "      for (var i = 0; i < script_attrs.length; i++) {\n", "        toinsert[toinsert.length - 1].firstChild.setAttribute(script_attrs[i].name, script_attrs[i].value);\n", "      }\n", "      // store reference to server id on output_area\n", "      output_area._bokeh_server_id = output.metadata[EXEC_MIME_TYPE][\"server_id\"];\n", "    }\n", "  }\n", "\n", "  function register_renderer(events, OutputArea) {\n", "\n", "    function append_mime(data, metadata, element) {\n", "      // create a DOM node to render to\n", "      var toinsert = this.create_output_subarea(\n", "        metadata,\n", "        CLASS_NAME,\n", "        EXEC_MIME_TYPE\n", "      );\n", "      this.keyboard_manager.register_events(toinsert);\n", "      // Render to node\n", "      var props = {data: data, metadata: metadata[EXEC_MIME_TYPE]};\n", "      render(props, toinsert[toinsert.length - 1]);\n", "      element.append(toinsert);\n", "      return toinsert\n", "    }\n", "\n", "    /* Handle when an output is cleared or removed */\n", "    events.on('clear_output.CodeCell', handleClearOutput);\n", "    events.on('delete.Cell', handleClearOutput);\n", "\n", "    /* Handle when a new output is added */\n", "    events.on('output_added.OutputArea', handleAddOutput);\n", "\n", "    /**\n", "     * Register the mime type and append_mime function with output_area\n", "     */\n", "    OutputArea.prototype.register_mime_type(EXEC_MIME_TYPE, append_mime, {\n", "      /* Is output safe? */\n", "      safe: true,\n", "      /* Index of renderer in `output_area.display_order` */\n", "      index: 0\n", "    });\n", "  }\n", "\n", "  // register the mime type if in Jupyter Notebook environment and previously unregistered\n", "  if (root.Jupyter !== undefined) {\n", "    var events = require('base/js/events');\n", "    var OutputArea = require('notebook/js/outputarea').OutputArea;\n", "\n", "    if (OutputArea.prototype.mime_types().indexOf(EXEC_MIME_TYPE) == -1) {\n", "      register_renderer(events, OutputArea);\n", "    }\n", "  }\n", "\n", "  \n", "  if (typeof (root._bokeh_timeout) === \"undefined\" || force === true) {\n", "    root._bokeh_timeout = Date.now() + 5000;\n", "    root._bokeh_failed_load = false;\n", "  }\n", "\n", "  var NB_LOAD_WARNING = {'data': {'text/html':\n", "     \"<div style='background-color: #fdd'>\\n\"+\n", "     \"<p>\\n\"+\n", "     \"BokehJS does not appear to have successfully loaded. If loading BokehJS from CDN, this \\n\"+\n", "     \"may be due to a slow or bad network connection. Possible fixes:\\n\"+\n", "     \"</p>\\n\"+\n", "     \"<ul>\\n\"+\n", "     \"<li>re-rerun `output_notebook()` to attempt to load from CDN again, or</li>\\n\"+\n", "     \"<li>use INLINE resources instead, as so:</li>\\n\"+\n", "     \"</ul>\\n\"+\n", "     \"<code>\\n\"+\n", "     \"from bokeh.resources import INLINE\\n\"+\n", "     \"output_notebook(resources=INLINE)\\n\"+\n", "     \"</code>\\n\"+\n", "     \"</div>\"}};\n", "\n", "  function display_loaded() {\n", "    var el = document.getElementById(\"1001\");\n", "    if (el != null) {\n", "      el.textContent = \"BokehJS is loading...\";\n", "    }\n", "    if (root.Bokeh !== undefined) {\n", "      if (el != null) {\n", "        el.textContent = \"BokehJS \" + root.Bokeh.version + \" successfully loaded.\";\n", "      }\n", "    } else if (Date.now() < root._bokeh_timeout) {\n", "      setTimeout(display_loaded, 100)\n", "    }\n", "  }\n", "\n", "\n", "  function run_callbacks() {\n", "    try {\n", "      root._bokeh_onload_callbacks.forEach(function(callback) {\n", "        if (callback != null)\n", "          callback();\n", "      });\n", "    } finally {\n", "      delete root._bokeh_onload_callbacks\n", "    }\n", "    console.debug(\"Bokeh: all callbacks have finished\");\n", "  }\n", "\n", "  function load_libs(css_urls, js_urls, callback) {\n", "    if (css_urls == null) css_urls = [];\n", "    if (js_urls == null) js_urls = [];\n", "\n", "    root._bokeh_onload_callbacks.push(callback);\n", "    if (root._bokeh_is_loading > 0) {\n", "      console.debug(\"Bokeh: BokehJS is being loaded, scheduling callback at\", now());\n", "      return null;\n", "    }\n", "    if (js_urls == null || js_urls.length === 0) {\n", "      run_callbacks();\n", "      return null;\n", "    }\n", "    console.debug(\"Bokeh: BokehJS not loaded, scheduling load and callback at\", now());\n", "    root._bokeh_is_loading = css_urls.length + js_urls.length;\n", "\n", "    function on_load() {\n", "      root._bokeh_is_loading--;\n", "      if (root._bokeh_is_loading === 0) {\n", "        console.debug(\"Bokeh: all BokehJS libraries/stylesheets loaded\");\n", "        run_callbacks()\n", "      }\n", "    }\n", "\n", "    function on_error() {\n", "      console.error(\"failed to load \" + url);\n", "    }\n", "\n", "    for (var i = 0; i < css_urls.length; i++) {\n", "      var url = css_urls[i];\n", "      const element = document.createElement(\"link\");\n", "      element.onload = on_load;\n", "      element.onerror = on_error;\n", "      element.rel = \"stylesheet\";\n", "      element.type = \"text/css\";\n", "      element.href = url;\n", "      console.debug(\"Bokeh: injecting link tag for BokehJS stylesheet: \", url);\n", "      document.body.appendChild(element);\n", "    }\n", "\n", "    for (var i = 0; i < js_urls.length; i++) {\n", "      var url = js_urls[i];\n", "      var element = document.createElement('script');\n", "      element.onload = on_load;\n", "      element.onerror = on_error;\n", "      element.async = false;\n", "      element.src = url;\n", "      console.debug(\"Bokeh: injecting script tag for BokehJS library: \", url);\n", "      document.head.appendChild(element);\n", "    }\n", "  };var element = document.getElementById(\"1001\");\n", "  if (element == null) {\n", "    console.error(\"Bokeh: ERROR: autoload.js configured with elementid '1001' but no matching script tag was found. \")\n", "    return false;\n", "  }\n", "\n", "  function inject_raw_css(css) {\n", "    const element = document.createElement(\"style\");\n", "    element.appendChild(document.createTextNode(css));\n", "    document.body.appendChild(element);\n", "  }\n", "\n", "  \n", "  var js_urls = [\"https://cdn.pydata.org/bokeh/release/bokeh-1.4.0.min.js\", \"https://cdn.pydata.org/bokeh/release/bokeh-widgets-1.4.0.min.js\", \"https://cdn.pydata.org/bokeh/release/bokeh-tables-1.4.0.min.js\", \"https://cdn.pydata.org/bokeh/release/bokeh-gl-1.4.0.min.js\"];\n", "  var css_urls = [];\n", "  \n", "\n", "  var inline_js = [\n", "    function(Bokeh) {\n", "      Bokeh.set_log_level(\"info\");\n", "    },\n", "    function(Bokeh) {\n", "    \n", "    \n", "    }\n", "  ];\n", "\n", "  function run_inline_js() {\n", "    \n", "    if (root.Bokeh !== undefined || force === true) {\n", "      \n", "    for (var i = 0; i < inline_js.length; i++) {\n", "      inline_js[i].call(root, root.Bokeh);\n", "    }\n", "    if (force === true) {\n", "        display_loaded();\n", "      }} else if (Date.now() < root._bokeh_timeout) {\n", "      setTimeout(run_inline_js, 100);\n", "    } else if (!root._bokeh_failed_load) {\n", "      console.log(\"Bokeh: BokehJS failed to load within specified timeout.\");\n", "      root._bokeh_failed_load = true;\n", "    } else if (force !== true) {\n", "      var cell = $(document.getElementById(\"1001\")).parents('.cell').data().cell;\n", "      cell.output_area.append_execute_result(NB_LOAD_WARNING)\n", "    }\n", "\n", "  }\n", "\n", "  if (root._bokeh_is_loading === 0) {\n", "    console.debug(\"Bokeh: BokehJS loaded, going straight to plotting\");\n", "    run_inline_js();\n", "  } else {\n", "    load_libs(css_urls, js_urls, function() {\n", "      console.debug(\"Bokeh: BokehJS plotting callback run at\", now());\n", "      run_inline_js();\n", "    });\n", "  }\n", "}(window));"], "application/vnd.bokehjs_load.v0+json": "\n(function(root) {\n  function now() {\n    return new Date();\n  }\n\n  var force = true;\n\n  if (typeof root._bokeh_onload_callbacks === \"undefined\" || force === true) {\n    root._bokeh_onload_callbacks = [];\n    root._bokeh_is_loading = undefined;\n  }\n\n  \n\n  \n  if (typeof (root._bokeh_timeout) === \"undefined\" || force === true) {\n    root._bokeh_timeout = Date.now() + 5000;\n    root._bokeh_failed_load = false;\n  }\n\n  var NB_LOAD_WARNING = {'data': {'text/html':\n     \"<div style='background-color: #fdd'>\\n\"+\n     \"<p>\\n\"+\n     \"BokehJS does not appear to have successfully loaded. If loading BokehJS from CDN, this \\n\"+\n     \"may be due to a slow or bad network connection. Possible fixes:\\n\"+\n     \"</p>\\n\"+\n     \"<ul>\\n\"+\n     \"<li>re-rerun `output_notebook()` to attempt to load from CDN again, or</li>\\n\"+\n     \"<li>use INLINE resources instead, as so:</li>\\n\"+\n     \"</ul>\\n\"+\n     \"<code>\\n\"+\n     \"from bokeh.resources import INLINE\\n\"+\n     \"output_notebook(resources=INLINE)\\n\"+\n     \"</code>\\n\"+\n     \"</div>\"}};\n\n  function display_loaded() {\n    var el = document.getElementById(\"1001\");\n    if (el != null) {\n      el.textContent = \"BokehJS is loading...\";\n    }\n    if (root.Bokeh !== undefined) {\n      if (el != null) {\n        el.textContent = \"BokehJS \" + root.Bokeh.version + \" successfully loaded.\";\n      }\n    } else if (Date.now() < root._bokeh_timeout) {\n      setTimeout(display_loaded, 100)\n    }\n  }\n\n\n  function run_callbacks() {\n    try {\n      root._bokeh_onload_callbacks.forEach(function(callback) {\n        if (callback != null)\n          callback();\n      });\n    } finally {\n      delete root._bokeh_onload_callbacks\n    }\n    console.debug(\"Bokeh: all callbacks have finished\");\n  }\n\n  function load_libs(css_urls, js_urls, callback) {\n    if (css_urls == null) css_urls = [];\n    if (js_urls == null) js_urls = [];\n\n    root._bokeh_onload_callbacks.push(callback);\n    if (root._bokeh_is_loading > 0) {\n      console.debug(\"Bokeh: BokehJS is being loaded, scheduling callback at\", now());\n      return null;\n    }\n    if (js_urls == null || js_urls.length === 0) {\n      run_callbacks();\n      return null;\n    }\n    console.debug(\"Bokeh: BokehJS not loaded, scheduling load and callback at\", now());\n    root._bokeh_is_loading = css_urls.length + js_urls.length;\n\n    function on_load() {\n      root._bokeh_is_loading--;\n      if (root._bokeh_is_loading === 0) {\n        console.debug(\"Bokeh: all BokehJS libraries/stylesheets loaded\");\n        run_callbacks()\n      }\n    }\n\n    function on_error() {\n      console.error(\"failed to load \" + url);\n    }\n\n    for (var i = 0; i < css_urls.length; i++) {\n      var url = css_urls[i];\n      const element = document.createElement(\"link\");\n      element.onload = on_load;\n      element.onerror = on_error;\n      element.rel = \"stylesheet\";\n      element.type = \"text/css\";\n      element.href = url;\n      console.debug(\"Bokeh: injecting link tag for BokehJS stylesheet: \", url);\n      document.body.appendChild(element);\n    }\n\n    for (var i = 0; i < js_urls.length; i++) {\n      var url = js_urls[i];\n      var element = document.createElement('script');\n      element.onload = on_load;\n      element.onerror = on_error;\n      element.async = false;\n      element.src = url;\n      console.debug(\"Bokeh: injecting script tag for BokehJS library: \", url);\n      document.head.appendChild(element);\n    }\n  };var element = document.getElementById(\"1001\");\n  if (element == null) {\n    console.error(\"Bokeh: ERROR: autoload.js configured with elementid '1001' but no matching script tag was found. \")\n    return false;\n  }\n\n  function inject_raw_css(css) {\n    const element = document.createElement(\"style\");\n    element.appendChild(document.createTextNode(css));\n    document.body.appendChild(element);\n  }\n\n  \n  var js_urls = [\"https://cdn.pydata.org/bokeh/release/bokeh-1.4.0.min.js\", \"https://cdn.pydata.org/bokeh/release/bokeh-widgets-1.4.0.min.js\", \"https://cdn.pydata.org/bokeh/release/bokeh-tables-1.4.0.min.js\", \"https://cdn.pydata.org/bokeh/release/bokeh-gl-1.4.0.min.js\"];\n  var css_urls = [];\n  \n\n  var inline_js = [\n    function(Bokeh) {\n      Bokeh.set_log_level(\"info\");\n    },\n    function(Bokeh) {\n    \n    \n    }\n  ];\n\n  function run_inline_js() {\n    \n    if (root.Bokeh !== undefined || force === true) {\n      \n    for (var i = 0; i < inline_js.length; i++) {\n      inline_js[i].call(root, root.Bokeh);\n    }\n    if (force === true) {\n        display_loaded();\n      }} else if (Date.now() < root._bokeh_timeout) {\n      setTimeout(run_inline_js, 100);\n    } else if (!root._bokeh_failed_load) {\n      console.log(\"Bokeh: BokehJS failed to load within specified timeout.\");\n      root._bokeh_failed_load = true;\n    } else if (force !== true) {\n      var cell = $(document.getElementById(\"1001\")).parents('.cell').data().cell;\n      cell.output_area.append_execute_result(NB_LOAD_WARNING)\n    }\n\n  }\n\n  if (root._bokeh_is_loading === 0) {\n    console.debug(\"Bokeh: BokehJS loaded, going straight to plotting\");\n    run_inline_js();\n  } else {\n    load_libs(css_urls, js_urls, function() {\n      console.debug(\"Bokeh: BokehJS plotting callback run at\", now());\n      run_inline_js();\n    });\n  }\n}(window));"}, "metadata": {}, "output_type": "display_data"}], "source": ["import bokeh, bokeh.io as bio\n", "bio.output_notebook()"]}, {"cell_type": "code", "execution_count": 44, "metadata": {"scrolled": false}, "outputs": [{"data": {"text/html": ["\n", "\n", "\n", "\n", "\n", "\n", "  <div class=\"bk-root\" id=\"ea645201-cb7b-4b13-b5ac-791d7836e501\" data-root-id=\"1002\"></div>\n"]}, "metadata": {}, "output_type": "display_data"}, {"data": {"application/javascript": ["(function(root) {\n", "  function embed_document(root) {\n", "    \n", "  var docs_json = {\"098eb251-cf35-4579-bf7a-8f38cc60d5ee\":{\"roots\":{\"references\":[{\"attributes\":{\"below\":[{\"id\":\"1013\",\"type\":\"LinearAxis\"}],\"center\":[{\"id\":\"1017\",\"type\":\"Grid\"},{\"id\":\"1022\",\"type\":\"Grid\"}],\"left\":[{\"id\":\"1018\",\"type\":\"LinearAxis\"}],\"renderers\":[{\"id\":\"1039\",\"type\":\"GlyphRenderer\"},{\"id\":\"1044\",\"type\":\"GlyphRenderer\"}],\"title\":{\"id\":\"1003\",\"type\":\"Title\"},\"toolbar\":{\"id\":\"1029\",\"type\":\"Toolbar\"},\"x_range\":{\"id\":\"1005\",\"type\":\"DataRange1d\"},\"x_scale\":{\"id\":\"1009\",\"type\":\"LinearScale\"},\"y_range\":{\"id\":\"1007\",\"type\":\"DataRange1d\"},\"y_scale\":{\"id\":\"1011\",\"type\":\"LinearScale\"}},\"id\":\"1002\",\"subtype\":\"Figure\",\"type\":\"Plot\"},{\"attributes\":{},\"id\":\"1019\",\"type\":\"BasicTicker\"},{\"attributes\":{\"source\":{\"id\":\"1041\",\"type\":\"ColumnDataSource\"}},\"id\":\"1045\",\"type\":\"CDSView\"},{\"attributes\":{\"dimension\":1,\"ticker\":{\"id\":\"1019\",\"type\":\"BasicTicker\"}},\"id\":\"1022\",\"type\":\"Grid\"},{\"attributes\":{},\"id\":\"1048\",\"type\":\"BasicTickFormatter\"},{\"attributes\":{\"callback\":null},\"id\":\"1005\",\"type\":\"DataRange1d\"},{\"attributes\":{\"fill_color\":{\"value\":\"#1f77b4\"},\"line_color\":{\"value\":\"#1f77b4\"},\"x\":{\"field\":\"x\"},\"y\":{\"field\":\"y\"}},\"id\":\"1037\",\"type\":\"Circle\"},{\"attributes\":{},\"id\":\"1050\",\"type\":\"BasicTickFormatter\"},{\"attributes\":{\"callback\":null},\"id\":\"1007\",\"type\":\"DataRange1d\"},{\"attributes\":{},\"id\":\"1051\",\"type\":\"Selection\"},{\"attributes\":{\"callback\":null,\"data\":{\"x\":{\"__ndarray__\":\"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\",\"dtype\":\"float64\",\"shape\":[107]},\"y\":{\"__ndarray__\":\"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\",\"dtype\":\"float64\",\"shape\":[107]}},\"selected\":{\"id\":\"1051\",\"type\":\"Selection\"},\"selection_policy\":{\"id\":\"1052\",\"type\":\"UnionRenderers\"}},\"id\":\"1036\",\"type\":\"ColumnDataSource\"},{\"attributes\":{},\"id\":\"1023\",\"type\":\"PanTool\"},{\"attributes\":{},\"id\":\"1052\",\"type\":\"UnionRenderers\"},{\"attributes\":{},\"id\":\"1024\",\"type\":\"WheelZoomTool\"},{\"attributes\":{},\"id\":\"1053\",\"type\":\"Selection\"},{\"attributes\":{\"overlay\":{\"id\":\"1055\",\"type\":\"BoxAnnotation\"}},\"id\":\"1025\",\"type\":\"BoxZoomTool\"},{\"attributes\":{},\"id\":\"1054\",\"type\":\"UnionRenderers\"},{\"attributes\":{\"data_source\":{\"id\":\"1041\",\"type\":\"ColumnDataSource\"},\"glyph\":{\"id\":\"1042\",\"type\":\"Text\"},\"hover_glyph\":null,\"muted_glyph\":null,\"nonselection_glyph\":{\"id\":\"1043\",\"type\":\"Text\"},\"selection_glyph\":null,\"view\":{\"id\":\"1045\",\"type\":\"CDSView\"}},\"id\":\"1044\",\"type\":\"GlyphRenderer\"},{\"attributes\":{},\"id\":\"1026\",\"type\":\"SaveTool\"},{\"attributes\":{\"bottom_units\":\"screen\",\"fill_alpha\":{\"value\":0.5},\"fill_color\":{\"value\":\"lightgrey\"},\"left_units\":\"screen\",\"level\":\"overlay\",\"line_alpha\":{\"value\":1.0},\"line_color\":{\"value\":\"black\"},\"line_dash\":[4,4],\"line_width\":{\"value\":2},\"render_mode\":\"css\",\"right_units\":\"screen\",\"top_units\":\"screen\"},\"id\":\"1055\",\"type\":\"BoxAnnotation\"},{\"attributes\":{},\"id\":\"1027\",\"type\":\"ResetTool\"},{\"attributes\":{},\"id\":\"1028\",\"type\":\"HelpTool\"},{\"attributes\":{\"active_drag\":\"auto\",\"active_inspect\":\"auto\",\"active_multi\":null,\"active_scroll\":\"auto\",\"active_tap\":\"auto\",\"tools\":[{\"id\":\"1023\",\"type\":\"PanTool\"},{\"id\":\"1024\",\"type\":\"WheelZoomTool\"},{\"id\":\"1025\",\"type\":\"BoxZoomTool\"},{\"id\":\"1026\",\"type\":\"SaveTool\"},{\"id\":\"1027\",\"type\":\"ResetTool\"},{\"id\":\"1028\",\"type\":\"HelpTool\"}]},\"id\":\"1029\",\"type\":\"Toolbar\"},{\"attributes\":{\"fill_alpha\":{\"value\":0.1},\"fill_color\":{\"value\":\"#1f77b4\"},\"line_alpha\":{\"value\":0.1},\"line_color\":{\"value\":\"#1f77b4\"},\"x\":{\"field\":\"x\"},\"y\":{\"field\":\"y\"}},\"id\":\"1038\",\"type\":\"Circle\"},{\"attributes\":{},\"id\":\"1009\",\"type\":\"LinearScale\"},{\"attributes\":{\"data_source\":{\"id\":\"1036\",\"type\":\"ColumnDataSource\"},\"glyph\":{\"id\":\"1037\",\"type\":\"Circle\"},\"hover_glyph\":null,\"muted_glyph\":null,\"nonselection_glyph\":{\"id\":\"1038\",\"type\":\"Circle\"},\"selection_glyph\":null,\"view\":{\"id\":\"1040\",\"type\":\"CDSView\"}},\"id\":\"1039\",\"type\":\"GlyphRenderer\"},{\"attributes\":{\"source\":{\"id\":\"1036\",\"type\":\"ColumnDataSource\"}},\"id\":\"1040\",\"type\":\"CDSView\"},{\"attributes\":{},\"id\":\"1011\",\"type\":\"LinearScale\"},{\"attributes\":{\"formatter\":{\"id\":\"1050\",\"type\":\"BasicTickFormatter\"},\"ticker\":{\"id\":\"1014\",\"type\":\"BasicTicker\"}},\"id\":\"1013\",\"type\":\"LinearAxis\"},{\"attributes\":{\"angle\":{\"units\":\"rad\",\"value\":0.1},\"text_baseline\":\"middle\",\"text_color\":{\"value\":\"black\"},\"text_font_size\":{\"value\":\"8pt\"},\"x\":{\"field\":\"x\"},\"y\":{\"field\":\"y\"}},\"id\":\"1042\",\"type\":\"Text\"},{\"attributes\":{},\"id\":\"1014\",\"type\":\"BasicTicker\"},{\"attributes\":{\"angle\":{\"units\":\"rad\",\"value\":0.1},\"text_alpha\":{\"value\":0.1},\"text_baseline\":\"middle\",\"text_color\":{\"value\":\"black\"},\"text_font_size\":{\"value\":\"8pt\"},\"x\":{\"field\":\"x\"},\"y\":{\"field\":\"y\"}},\"id\":\"1043\",\"type\":\"Text\"},{\"attributes\":{\"callback\":null,\"data\":{\"text\":[\"age\",\"workclass__Self-emp-not-inc\",\"workclass__Private\",\"workclass__Federal-gov\",\"workclass__Local-gov\",\"workclass__?\",\"workclass__Self-emp-inc\",\"workclass__Without-pay\",\"workclass__Never-worked\",\"workclass__nan\",\"education__HS-grad\",\"education__11th\",\"education__Masters\",\"education__9th\",\"education__Some-college\",\"education__Assoc-acdm\",\"education__Assoc-voc\",\"education__7th-8th\",\"education__Doctorate\",\"education__Prof-school\",\"education__5th-6th\",\"education__10th\",\"education__1st-4th\",\"education__Preschool\",\"education__12th\",\"education__nan\",\"education_num\",\"marital_status__Married-civ-spouse\",\"marital_status__Divorced\",\"marital_status__Married-spouse-absent\",\"marital_status__Separated\",\"marital_status__Married-AF-spouse\",\"marital_status__Widowed\",\"marital_status__nan\",\"occupation__Exec-managerial\",\"occupation__Handlers-cleaners\",\"occupation__Prof-specialty\",\"occupation__Other-service\",\"occupation__Sales\",\"occupation__Craft-repair\",\"occupation__Transport-moving\",\"occupation__Farming-fishing\",\"occupation__Machine-op-inspct\",\"occupation__Tech-support\",\"occupation__?\",\"occupation__Protective-serv\",\"occupation__Armed-Forces\",\"occupation__Priv-house-serv\",\"occupation__nan\",\"relationship__Husband\",\"relationship__Wife\",\"relationship__Own-child\",\"relationship__Unmarried\",\"relationship__Other-relative\",\"relationship__nan\",\"race__Black\",\"race__Asian-Pac-Islander\",\"race__Amer-Indian-Eskimo\",\"race__Other\",\"race__nan\",\"sex__Female\",\"sex__nan\",\"capital_gain\",\"capital_loss\",\"hours_per_week\",\"native_country__Cuba\",\"native_country__Jamaica\",\"native_country__India\",\"native_country__?\",\"native_country__Mexico\",\"native_country__South\",\"native_country__Puerto-Rico\",\"native_country__Honduras\",\"native_country__England\",\"native_country__Canada\",\"native_country__Germany\",\"native_country__Iran\",\"native_country__Philippines\",\"native_country__Italy\",\"native_country__Poland\",\"native_country__Columbia\",\"native_country__Cambodia\",\"native_country__Thailand\",\"native_country__Ecuador\",\"native_country__Laos\",\"native_country__Taiwan\",\"native_country__Haiti\",\"native_country__Portugal\",\"native_country__Dominican-Republic\",\"native_country__El-Salvador\",\"native_country__France\",\"native_country__Guatemala\",\"native_country__China\",\"native_country__Japan\",\"native_country__Yugoslavia\",\"native_country__Peru\",\"native_country__Outlying-US(Guam-USVI-etc)\",\"native_country__Scotland\",\"native_country__Trinadad&Tobago\",\"native_country__Greece\",\"native_country__Nicaragua\",\"native_country__Vietnam\",\"native_country__Hong\",\"native_country__Ireland\",\"native_country__Hungary\",\"native_country__Holand-Netherlands\",\"native_country__nan\"],\"x\":{\"__ndarray__\":\"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\",\"dtype\":\"float64\",\"shape\":[107]},\"y\":{\"__ndarray__\":\"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\",\"dtype\":\"float64\",\"shape\":[107]}},\"selected\":{\"id\":\"1053\",\"type\":\"Selection\"},\"selection_policy\":{\"id\":\"1054\",\"type\":\"UnionRenderers\"}},\"id\":\"1041\",\"type\":\"ColumnDataSource\"},{\"attributes\":{\"ticker\":{\"id\":\"1014\",\"type\":\"BasicTicker\"}},\"id\":\"1017\",\"type\":\"Grid\"},{\"attributes\":{\"text\":\"ACP sur les variables - axe 1, 2\"},\"id\":\"1003\",\"type\":\"Title\"},{\"attributes\":{\"formatter\":{\"id\":\"1048\",\"type\":\"BasicTickFormatter\"},\"ticker\":{\"id\":\"1019\",\"type\":\"BasicTicker\"}},\"id\":\"1018\",\"type\":\"LinearAxis\"}],\"root_ids\":[\"1002\"]},\"title\":\"Bokeh Application\",\"version\":\"1.4.0\"}};\n", "  var render_items = [{\"docid\":\"098eb251-cf35-4579-bf7a-8f38cc60d5ee\",\"roots\":{\"1002\":\"ea645201-cb7b-4b13-b5ac-791d7836e501\"}}];\n", "  root.Bokeh.embed.embed_items_notebook(docs_json, render_items);\n", "\n", "  }\n", "  if (root.Bokeh !== undefined) {\n", "    embed_document(root);\n", "  } else {\n", "    var attempts = 0;\n", "    var timer = setInterval(function(root) {\n", "      if (root.Bokeh !== undefined) {\n", "        clearInterval(timer);\n", "        embed_document(root);\n", "      } else {\n", "        attempts++;\n", "        if (attempts > 100) {\n", "          clearInterval(timer);\n", "          console.log(\"Bokeh: ERROR: Unable to run BokehJS code because BokehJS library is missing\");\n", "        }\n", "      }\n", "    }, 10, root)\n", "  }\n", "})(window);"], "application/vnd.bokehjs_exec.v0+json": ""}, "metadata": {"application/vnd.bokehjs_exec.v0+json": {"id": "1002"}}, "output_type": "display_data"}], "source": ["from bokeh.plotting import figure, show\n", "p = figure(title=\"ACP sur les variables - axe 1, 2\")\n", "p.circle(tr[\"axe1\"], tr[\"axe2\"])\n", "p.text(tr[\"axe1\"], tr[\"axe2\"], tr.index,\n", "       text_font_size=\"8pt\", text_baseline=\"middle\", angle=0.1)\n", "show(p)"]}, {"cell_type": "markdown", "metadata": {}, "source": ["## Analyse d'erreur\n", "\n", "On recherche les erreurs les plus flagrantes, celles dont le score est \u00e9lev\u00e9."]}, {"cell_type": "code", "execution_count": 45, "metadata": {}, "outputs": [], "source": ["pred = pipe.predict(X_test)\n", "proba = pipe.predict_proba(X_test)\n", "pred2 = pipe2.predict(X_test)\n", "proba2 = pipe2.predict_proba(X_test)\n", "pred3 = pipe3.predict(X_test)\n", "proba3 = pipe3.predict_proba(X_test)\n", "pred4 = pipe4.predict(X_test)\n", "proba4 = pipe4.predict_proba(X_test)"]}, {"cell_type": "code", "execution_count": 46, "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_test</th>\n", "      <th>pred1</th>\n", "      <th>P1(&gt;=50K)</th>\n", "      <th>pred2</th>\n", "      <th>P2(&gt;=50K)</th>\n", "      <th>pred3</th>\n", "      <th>P3(&gt;=50K)</th>\n", "      <th>pred4</th>\n", "      <th>P4(&gt;=50K)</th>\n", "      <th>age</th>\n", "      <th>...</th>\n", "      <th>education_num</th>\n", "      <th>marital_status</th>\n", "      <th>occupation</th>\n", "      <th>relationship</th>\n", "      <th>race</th>\n", "      <th>sex</th>\n", "      <th>capital_gain</th>\n", "      <th>capital_loss</th>\n", "      <th>hours_per_week</th>\n", "      <th>native_country</th>\n", "    </tr>\n", "  </thead>\n", "  <tbody>\n", "    <tr>\n", "      <th>0</th>\n", "      <td>0.0</td>\n", "      <td>0.0</td>\n", "      <td>0.010831</td>\n", "      <td>0.0</td>\n", "      <td>0.000000</td>\n", "      <td>0.0</td>\n", "      <td>0.007642</td>\n", "      <td>0.0</td>\n", "      <td>0.000384</td>\n", "      <td>25</td>\n", "      <td>...</td>\n", "      <td>7</td>\n", "      <td>Never-married</td>\n", "      <td>Machine-op-inspct</td>\n", "      <td>Own-child</td>\n", "      <td>Black</td>\n", "      <td>Male</td>\n", "      <td>0</td>\n", "      <td>0</td>\n", "      <td>40</td>\n", "      <td>United-States</td>\n", "    </tr>\n", "    <tr>\n", "      <th>1</th>\n", "      <td>0.0</td>\n", "      <td>0.0</td>\n", "      <td>0.179896</td>\n", "      <td>0.0</td>\n", "      <td>0.030000</td>\n", "      <td>0.0</td>\n", "      <td>0.203497</td>\n", "      <td>0.0</td>\n", "      <td>0.207168</td>\n", "      <td>38</td>\n", "      <td>...</td>\n", "      <td>9</td>\n", "      <td>Married-civ-spouse</td>\n", "      <td>Farming-fishing</td>\n", "      <td>Husband</td>\n", "      <td>White</td>\n", "      <td>Male</td>\n", "      <td>0</td>\n", "      <td>0</td>\n", "      <td>50</td>\n", "      <td>United-States</td>\n", "    </tr>\n", "    <tr>\n", "      <th>2</th>\n", "      <td>1.0</td>\n", "      <td>1.0</td>\n", "      <td>0.526712</td>\n", "      <td>1.0</td>\n", "      <td>0.526667</td>\n", "      <td>0.0</td>\n", "      <td>0.277167</td>\n", "      <td>0.0</td>\n", "      <td>0.471523</td>\n", "      <td>28</td>\n", "      <td>...</td>\n", "      <td>12</td>\n", "      <td>Married-civ-spouse</td>\n", "      <td>Protective-serv</td>\n", "      <td>Husband</td>\n", "      <td>White</td>\n", "      <td>Male</td>\n", "      <td>0</td>\n", "      <td>0</td>\n", "      <td>40</td>\n", "      <td>United-States</td>\n", "    </tr>\n", "    <tr>\n", "      <th>3</th>\n", "      <td>1.0</td>\n", "      <td>1.0</td>\n", "      <td>0.789027</td>\n", "      <td>1.0</td>\n", "      <td>0.930000</td>\n", "      <td>1.0</td>\n", "      <td>0.984138</td>\n", "      <td>1.0</td>\n", "      <td>0.987516</td>\n", "      <td>44</td>\n", "      <td>...</td>\n", "      <td>10</td>\n", "      <td>Married-civ-spouse</td>\n", "      <td>Machine-op-inspct</td>\n", "      <td>Husband</td>\n", "      <td>Black</td>\n", "      <td>Male</td>\n", "      <td>7688</td>\n", "      <td>0</td>\n", "      <td>40</td>\n", "      <td>United-States</td>\n", "    </tr>\n", "    <tr>\n", "      <th>4</th>\n", "      <td>0.0</td>\n", "      <td>0.0</td>\n", "      <td>0.009680</td>\n", "      <td>0.0</td>\n", "      <td>0.000000</td>\n", "      <td>0.0</td>\n", "      <td>0.002210</td>\n", "      <td>0.0</td>\n", "      <td>0.000151</td>\n", "      <td>18</td>\n", "      <td>...</td>\n", "      <td>10</td>\n", "      <td>Never-married</td>\n", "      <td>?</td>\n", "      <td>Own-child</td>\n", "      <td>White</td>\n", "      <td>Female</td>\n", "      <td>0</td>\n", "      <td>0</td>\n", "      <td>30</td>\n", "      <td>United-States</td>\n", "    </tr>\n", "  </tbody>\n", "</table>\n", "<p>5 rows \u00d7 22 columns</p>\n", "</div>"], "text/plain": ["   y_test  pred1  P1(>=50K)  pred2  P2(>=50K)  pred3  P3(>=50K)  pred4  \\\n", "0     0.0    0.0   0.010831    0.0   0.000000    0.0   0.007642    0.0   \n", "1     0.0    0.0   0.179896    0.0   0.030000    0.0   0.203497    0.0   \n", "2     1.0    1.0   0.526712    1.0   0.526667    0.0   0.277167    0.0   \n", "3     1.0    1.0   0.789027    1.0   0.930000    1.0   0.984138    1.0   \n", "4     0.0    0.0   0.009680    0.0   0.000000    0.0   0.002210    0.0   \n", "\n", "   P4(>=50K)  age  ... education_num      marital_status         occupation  \\\n", "0   0.000384   25  ...             7       Never-married  Machine-op-inspct   \n", "1   0.207168   38  ...             9  Married-civ-spouse    Farming-fishing   \n", "2   0.471523   28  ...            12  Married-civ-spouse    Protective-serv   \n", "3   0.987516   44  ...            10  Married-civ-spouse  Machine-op-inspct   \n", "4   0.000151   18  ...            10       Never-married                  ?   \n", "\n", "  relationship   race     sex capital_gain capital_loss  hours_per_week  \\\n", "0    Own-child  Black    Male            0            0              40   \n", "1      Husband  White    Male            0            0              50   \n", "2      Husband  White    Male            0            0              40   \n", "3      Husband  Black    Male         7688            0              40   \n", "4    Own-child  White  Female            0            0              30   \n", "\n", "   native_country  \n", "0   United-States  \n", "1   United-States  \n", "2   United-States  \n", "3   United-States  \n", "4   United-States  \n", "\n", "[5 rows x 22 columns]"]}, "execution_count": 47, "metadata": {}, "output_type": "execute_result"}], "source": ["data = pandas.concat([\n", "            pandas.DataFrame(y_test.astype(float).values, columns=['y_test']),\n", "            pandas.DataFrame(pred, columns=['pred1']),\n", "            pandas.DataFrame(proba[:,1], columns=['P1(>=50K)']),\n", "            pandas.DataFrame(pred2, columns=['pred2']),\n", "            pandas.DataFrame(proba2[:,1], columns=['P2(>=50K)']),\n", "            pandas.DataFrame(pred3, columns=['pred3']),\n", "            pandas.DataFrame(proba3[:,1], columns=['P3(>=50K)']),\n", "            pandas.DataFrame(pred4, columns=['pred4']),\n", "            pandas.DataFrame(proba4[:,1], columns=['P4(>=50K)']),\n", "            X_test,\n", "            ], axis=1)\n", "data.head()"]}, {"cell_type": "code", "execution_count": 47, "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>3605</th>\n", "      <th>2926</th>\n", "      <th>13783</th>\n", "      <th>2247</th>\n", "      <th>13128</th>\n", "    </tr>\n", "  </thead>\n", "  <tbody>\n", "    <tr>\n", "      <th>y_test</th>\n", "      <td>0</td>\n", "      <td>0</td>\n", "      <td>0</td>\n", "      <td>0</td>\n", "      <td>0</td>\n", "    </tr>\n", "    <tr>\n", "      <th>pred1</th>\n", "      <td>1</td>\n", "      <td>1</td>\n", "      <td>1</td>\n", "      <td>1</td>\n", "      <td>1</td>\n", "    </tr>\n", "    <tr>\n", "      <th>P1(&gt;=50K)</th>\n", "      <td>1</td>\n", "      <td>0.793878</td>\n", "      <td>0.764859</td>\n", "      <td>0.788413</td>\n", "      <td>0.832539</td>\n", "    </tr>\n", "    <tr>\n", "      <th>pred2</th>\n", "      <td>1</td>\n", "      <td>1</td>\n", "      <td>1</td>\n", "      <td>1</td>\n", "      <td>1</td>\n", "    </tr>\n", "    <tr>\n", "      <th>P2(&gt;=50K)</th>\n", "      <td>0.74</td>\n", "      <td>0.74</td>\n", "      <td>0.937333</td>\n", "      <td>0.94</td>\n", "      <td>0.65</td>\n", "    </tr>\n", "    <tr>\n", "      <th>pred3</th>\n", "      <td>1</td>\n", "      <td>1</td>\n", "      <td>1</td>\n", "      <td>1</td>\n", "      <td>1</td>\n", "    </tr>\n", "    <tr>\n", "      <th>P3(&gt;=50K)</th>\n", "      <td>0.972606</td>\n", "      <td>0.900088</td>\n", "      <td>0.933867</td>\n", "      <td>0.796695</td>\n", "      <td>0.913922</td>\n", "    </tr>\n", "    <tr>\n", "      <th>pred4</th>\n", "      <td>1</td>\n", "      <td>1</td>\n", "      <td>1</td>\n", "      <td>1</td>\n", "      <td>1</td>\n", "    </tr>\n", "    <tr>\n", "      <th>P4(&gt;=50K)</th>\n", "      <td>0.998917</td>\n", "      <td>0.989239</td>\n", "      <td>0.980988</td>\n", "      <td>0.944513</td>\n", "      <td>0.931411</td>\n", "    </tr>\n", "    <tr>\n", "      <th>age</th>\n", "      <td>36</td>\n", "      <td>65</td>\n", "      <td>51</td>\n", "      <td>55</td>\n", "      <td>48</td>\n", "    </tr>\n", "    <tr>\n", "      <th>workclass</th>\n", "      <td>Self-emp-not-inc</td>\n", "      <td>Self-emp-not-inc</td>\n", "      <td>Private</td>\n", "      <td>Self-emp-inc</td>\n", "      <td>Local-gov</td>\n", "    </tr>\n", "    <tr>\n", "      <th>education</th>\n", "      <td>HS-grad</td>\n", "      <td>Masters</td>\n", "      <td>Some-college</td>\n", "      <td>Prof-school</td>\n", "      <td>Bachelors</td>\n", "    </tr>\n", "    <tr>\n", "      <th>education_num</th>\n", "      <td>9</td>\n", "      <td>14</td>\n", "      <td>10</td>\n", "      <td>15</td>\n", "      <td>13</td>\n", "    </tr>\n", "    <tr>\n", "      <th>marital_status</th>\n", "      <td>Married-civ-spouse</td>\n", "      <td>Married-spouse-absent</td>\n", "      <td>Married-civ-spouse</td>\n", "      <td>Married-civ-spouse</td>\n", "      <td>Separated</td>\n", "    </tr>\n", "    <tr>\n", "      <th>occupation</th>\n", "      <td>Exec-managerial</td>\n", "      <td>Prof-specialty</td>\n", "      <td>Exec-managerial</td>\n", "      <td>Prof-specialty</td>\n", "      <td>Prof-specialty</td>\n", "    </tr>\n", "    <tr>\n", "      <th>relationship</th>\n", "      <td>Husband</td>\n", "      <td>Not-in-family</td>\n", "      <td>Husband</td>\n", "      <td>Husband</td>\n", "      <td>Unmarried</td>\n", "    </tr>\n", "    <tr>\n", "      <th>race</th>\n", "      <td>Asian-Pac-Islander</td>\n", "      <td>White</td>\n", "      <td>White</td>\n", "      <td>White</td>\n", "      <td>White</td>\n", "    </tr>\n", "    <tr>\n", "      <th>sex</th>\n", "      <td>Male</td>\n", "      <td>Female</td>\n", "      <td>Male</td>\n", "      <td>Male</td>\n", "      <td>Female</td>\n", "    </tr>\n", "    <tr>\n", "      <th>capital_gain</th>\n", "      <td>41310</td>\n", "      <td>7978</td>\n", "      <td>0</td>\n", "      <td>0</td>\n", "      <td>7443</td>\n", "    </tr>\n", "    <tr>\n", "      <th>capital_loss</th>\n", "      <td>0</td>\n", "      <td>0</td>\n", "      <td>1902</td>\n", "      <td>0</td>\n", "      <td>0</td>\n", "    </tr>\n", "    <tr>\n", "      <th>hours_per_week</th>\n", "      <td>90</td>\n", "      <td>40</td>\n", "      <td>40</td>\n", "      <td>55</td>\n", "      <td>45</td>\n", "    </tr>\n", "    <tr>\n", "      <th>native_country</th>\n", "      <td>South</td>\n", "      <td>United-States</td>\n", "      <td>United-States</td>\n", "      <td>United-States</td>\n", "      <td>United-States</td>\n", "    </tr>\n", "  </tbody>\n", "</table>\n", "</div>"], "text/plain": ["                             3605                   2926                13783  \\\n", "y_test                           0                      0                   0   \n", "pred1                            1                      1                   1   \n", "P1(>=50K)                        1               0.793878            0.764859   \n", "pred2                            1                      1                   1   \n", "P2(>=50K)                     0.74                   0.74            0.937333   \n", "pred3                            1                      1                   1   \n", "P3(>=50K)                 0.972606               0.900088            0.933867   \n", "pred4                            1                      1                   1   \n", "P4(>=50K)                 0.998917               0.989239            0.980988   \n", "age                             36                     65                  51   \n", "workclass         Self-emp-not-inc       Self-emp-not-inc             Private   \n", "education                  HS-grad                Masters        Some-college   \n", "education_num                    9                     14                  10   \n", "marital_status  Married-civ-spouse  Married-spouse-absent  Married-civ-spouse   \n", "occupation         Exec-managerial         Prof-specialty     Exec-managerial   \n", "relationship               Husband          Not-in-family             Husband   \n", "race            Asian-Pac-Islander                  White               White   \n", "sex                           Male                 Female                Male   \n", "capital_gain                 41310                   7978                   0   \n", "capital_loss                     0                      0                1902   \n", "hours_per_week                  90                     40                  40   \n", "native_country               South          United-States       United-States   \n", "\n", "                             2247            13128  \n", "y_test                           0               0  \n", "pred1                            1               1  \n", "P1(>=50K)                 0.788413        0.832539  \n", "pred2                            1               1  \n", "P2(>=50K)                     0.94            0.65  \n", "pred3                            1               1  \n", "P3(>=50K)                 0.796695        0.913922  \n", "pred4                            1               1  \n", "P4(>=50K)                 0.944513        0.931411  \n", "age                             55              48  \n", "workclass             Self-emp-inc       Local-gov  \n", "education              Prof-school       Bachelors  \n", "education_num                   15              13  \n", "marital_status  Married-civ-spouse       Separated  \n", "occupation          Prof-specialty  Prof-specialty  \n", "relationship               Husband       Unmarried  \n", "race                         White           White  \n", "sex                           Male          Female  \n", "capital_gain                     0            7443  \n", "capital_loss                     0               0  \n", "hours_per_week                  55              45  \n", "native_country       United-States   United-States  "]}, "execution_count": 48, "metadata": {}, "output_type": "execute_result"}], "source": ["data[data.y_test != data.pred4].sort_values('P4(>=50K)', ascending=False).head().T"]}, {"cell_type": "markdown", "metadata": {}, "source": ["Tous les mod\u00e8les font l'erreur sur ces cinq exemples. Le mod\u00e8le a toutes les raisons de d\u00e9cider que les personnes gagnent plus de 50k par an, beaucoup d'\u00e9tudes, plut\u00f4t \u00e2ge ou travaillant beaucoup."]}, {"cell_type": "code", "execution_count": 48, "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>10408</th>\n", "      <th>5953</th>\n", "      <th>3059</th>\n", "    </tr>\n", "  </thead>\n", "  <tbody>\n", "    <tr>\n", "      <th>y_test</th>\n", "      <td>1</td>\n", "      <td>1</td>\n", "      <td>1</td>\n", "    </tr>\n", "    <tr>\n", "      <th>pred1</th>\n", "      <td>0</td>\n", "      <td>0</td>\n", "      <td>0</td>\n", "    </tr>\n", "    <tr>\n", "      <th>P1(&gt;=50K)</th>\n", "      <td>0.0146811</td>\n", "      <td>0.00972505</td>\n", "      <td>0.0145646</td>\n", "    </tr>\n", "    <tr>\n", "      <th>pred2</th>\n", "      <td>0</td>\n", "      <td>0</td>\n", "      <td>0</td>\n", "    </tr>\n", "    <tr>\n", "      <th>P2(&gt;=50K)</th>\n", "      <td>0</td>\n", "      <td>0</td>\n", "      <td>0</td>\n", "    </tr>\n", "    <tr>\n", "      <th>pred3</th>\n", "      <td>0</td>\n", "      <td>0</td>\n", "      <td>0</td>\n", "    </tr>\n", "    <tr>\n", "      <th>P3(&gt;=50K)</th>\n", "      <td>0.00288558</td>\n", "      <td>0.00299048</td>\n", "      <td>0.00561586</td>\n", "    </tr>\n", "    <tr>\n", "      <th>pred4</th>\n", "      <td>0</td>\n", "      <td>0</td>\n", "      <td>0</td>\n", "    </tr>\n", "    <tr>\n", "      <th>P4(&gt;=50K)</th>\n", "      <td>0.000149386</td>\n", "      <td>0.000395875</td>\n", "      <td>0.000863302</td>\n", "    </tr>\n", "    <tr>\n", "      <th>age</th>\n", "      <td>22</td>\n", "      <td>20</td>\n", "      <td>22</td>\n", "    </tr>\n", "    <tr>\n", "      <th>workclass</th>\n", "      <td>?</td>\n", "      <td>Private</td>\n", "      <td>Private</td>\n", "    </tr>\n", "    <tr>\n", "      <th>education</th>\n", "      <td>Some-college</td>\n", "      <td>12th</td>\n", "      <td>Some-college</td>\n", "    </tr>\n", "    <tr>\n", "      <th>education_num</th>\n", "      <td>10</td>\n", "      <td>8</td>\n", "      <td>10</td>\n", "    </tr>\n", "    <tr>\n", "      <th>marital_status</th>\n", "      <td>Never-married</td>\n", "      <td>Never-married</td>\n", "      <td>Never-married</td>\n", "    </tr>\n", "    <tr>\n", "      <th>occupation</th>\n", "      <td>?</td>\n", "      <td>Other-service</td>\n", "      <td>Sales</td>\n", "    </tr>\n", "    <tr>\n", "      <th>relationship</th>\n", "      <td>Own-child</td>\n", "      <td>Own-child</td>\n", "      <td>Not-in-family</td>\n", "    </tr>\n", "    <tr>\n", "      <th>race</th>\n", "      <td>White</td>\n", "      <td>Black</td>\n", "      <td>White</td>\n", "    </tr>\n", "    <tr>\n", "      <th>sex</th>\n", "      <td>Male</td>\n", "      <td>Male</td>\n", "      <td>Male</td>\n", "    </tr>\n", "    <tr>\n", "      <th>capital_gain</th>\n", "      <td>0</td>\n", "      <td>0</td>\n", "      <td>0</td>\n", "    </tr>\n", "    <tr>\n", "      <th>capital_loss</th>\n", "      <td>0</td>\n", "      <td>0</td>\n", "      <td>0</td>\n", "    </tr>\n", "    <tr>\n", "      <th>hours_per_week</th>\n", "      <td>15</td>\n", "      <td>35</td>\n", "      <td>25</td>\n", "    </tr>\n", "    <tr>\n", "      <th>native_country</th>\n", "      <td>?</td>\n", "      <td>United-States</td>\n", "      <td>United-States</td>\n", "    </tr>\n", "  </tbody>\n", "</table>\n", "</div>"], "text/plain": ["                        10408          5953           3059 \n", "y_test                      1              1              1\n", "pred1                       0              0              0\n", "P1(>=50K)           0.0146811     0.00972505      0.0145646\n", "pred2                       0              0              0\n", "P2(>=50K)                   0              0              0\n", "pred3                       0              0              0\n", "P3(>=50K)          0.00288558     0.00299048     0.00561586\n", "pred4                       0              0              0\n", "P4(>=50K)         0.000149386    0.000395875    0.000863302\n", "age                        22             20             22\n", "workclass                   ?        Private        Private\n", "education        Some-college           12th   Some-college\n", "education_num              10              8             10\n", "marital_status  Never-married  Never-married  Never-married\n", "occupation                  ?  Other-service          Sales\n", "relationship        Own-child      Own-child  Not-in-family\n", "race                    White          Black          White\n", "sex                      Male           Male           Male\n", "capital_gain                0              0              0\n", "capital_loss                0              0              0\n", "hours_per_week             15             35             25\n", "native_country              ?  United-States  United-States"]}, "execution_count": 49, "metadata": {}, "output_type": "execute_result"}], "source": ["wrong_study = data[data.y_test != data.pred4].sort_values('P4(>=50K)', ascending=True).head(n=3).T\n", "wrong_study"]}, {"cell_type": "markdown", "metadata": {}, "source": ["Ceux-ci sont probablement \u00e9tudiants et d\u00e9j\u00e0 ais\u00e9s. Il faudrait avoir quelques informations sur les parents pour confirmer. On recherche les plus proches voisins dans la base pour voir ce que le mod\u00e8le r\u00e9pond. Il faut n\u00e9anmoins appliquer cela sur la base une fois les cat\u00e9gories transform\u00e9es."]}, {"cell_type": "code", "execution_count": 49, "metadata": {}, "outputs": [{"data": {"text/plain": ["NearestNeighbors(algorithm='auto', leaf_size=30, metric='minkowski',\n", "                 metric_params=None, n_jobs=None, n_neighbors=5, p=2,\n", "                 radius=1.0)"]}, "execution_count": 50, "metadata": {}, "output_type": "execute_result"}], "source": ["from sklearn.neighbors import NearestNeighbors\n", "knn = NearestNeighbors()\n", "knn.fit(X_train_cat)"]}, {"cell_type": "code", "execution_count": 50, "metadata": {}, "outputs": [], "source": ["X_test_cat = pipe2.steps[0][-1].transform(X_test)\n", "X_test_cat.columns = X_train_cat.columns"]}, {"cell_type": "code", "execution_count": 51, "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_test</th>\n", "      <th>pred1</th>\n", "      <th>P1(&gt;=50K)</th>\n", "      <th>pred2</th>\n", "      <th>P2(&gt;=50K)</th>\n", "      <th>pred3</th>\n", "      <th>P3(&gt;=50K)</th>\n", "      <th>pred4</th>\n", "      <th>P4(&gt;=50K)</th>\n", "      <th>age</th>\n", "      <th>...</th>\n", "      <th>education_num</th>\n", "      <th>marital_status</th>\n", "      <th>occupation</th>\n", "      <th>relationship</th>\n", "      <th>race</th>\n", "      <th>sex</th>\n", "      <th>capital_gain</th>\n", "      <th>capital_loss</th>\n", "      <th>hours_per_week</th>\n", "      <th>native_country</th>\n", "    </tr>\n", "  </thead>\n", "  <tbody>\n", "    <tr>\n", "      <th>10408</th>\n", "      <td>1.0</td>\n", "      <td>0.0</td>\n", "      <td>0.014681</td>\n", "      <td>0.0</td>\n", "      <td>0.0</td>\n", "      <td>0.0</td>\n", "      <td>0.002886</td>\n", "      <td>0.0</td>\n", "      <td>0.000149</td>\n", "      <td>22</td>\n", "      <td>...</td>\n", "      <td>10</td>\n", "      <td>Never-married</td>\n", "      <td>?</td>\n", "      <td>Own-child</td>\n", "      <td>White</td>\n", "      <td>Male</td>\n", "      <td>0</td>\n", "      <td>0</td>\n", "      <td>15</td>\n", "      <td>?</td>\n", "    </tr>\n", "    <tr>\n", "      <th>5953</th>\n", "      <td>1.0</td>\n", "      <td>0.0</td>\n", "      <td>0.009725</td>\n", "      <td>0.0</td>\n", "      <td>0.0</td>\n", "      <td>0.0</td>\n", "      <td>0.002990</td>\n", "      <td>0.0</td>\n", "      <td>0.000396</td>\n", "      <td>20</td>\n", "      <td>...</td>\n", "      <td>8</td>\n", "      <td>Never-married</td>\n", "      <td>Other-service</td>\n", "      <td>Own-child</td>\n", "      <td>Black</td>\n", "      <td>Male</td>\n", "      <td>0</td>\n", "      <td>0</td>\n", "      <td>35</td>\n", "      <td>United-States</td>\n", "    </tr>\n", "  </tbody>\n", "</table>\n", "<p>2 rows \u00d7 22 columns</p>\n", "</div>"], "text/plain": ["       y_test  pred1  P1(>=50K)  pred2  P2(>=50K)  pred3  P3(>=50K)  pred4  \\\n", "10408     1.0    0.0   0.014681    0.0        0.0    0.0   0.002886    0.0   \n", "5953      1.0    0.0   0.009725    0.0        0.0    0.0   0.002990    0.0   \n", "\n", "       P4(>=50K)  age  ... education_num marital_status     occupation  \\\n", "10408   0.000149   22  ...            10  Never-married              ?   \n", "5953    0.000396   20  ...             8  Never-married  Other-service   \n", "\n", "      relationship   race   sex capital_gain capital_loss  hours_per_week  \\\n", "10408    Own-child  White  Male            0            0              15   \n", "5953     Own-child  Black  Male            0            0              35   \n", "\n", "       native_country  \n", "10408               ?  \n", "5953    United-States  \n", "\n", "[2 rows x 22 columns]"]}, "execution_count": 52, "metadata": {}, "output_type": "execute_result"}], "source": ["wrong = data[data.y_test != data.pred4].sort_values('P4(>=50K)', ascending=True).head()\n", "wrong.head(n=2)"]}, {"cell_type": "code", "execution_count": 52, "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>age</th>\n", "      <th>workclass__Self-emp-not-inc</th>\n", "      <th>workclass__Private</th>\n", "      <th>workclass__Federal-gov</th>\n", "      <th>workclass__Local-gov</th>\n", "      <th>workclass__?</th>\n", "      <th>workclass__Self-emp-inc</th>\n", "      <th>workclass__Without-pay</th>\n", "      <th>workclass__Never-worked</th>\n", "      <th>workclass__nan</th>\n", "      <th>...</th>\n", "      <th>native_country__Scotland</th>\n", "      <th>native_country__Trinadad&amp;Tobago</th>\n", "      <th>native_country__Greece</th>\n", "      <th>native_country__Nicaragua</th>\n", "      <th>native_country__Vietnam</th>\n", "      <th>native_country__Hong</th>\n", "      <th>native_country__Ireland</th>\n", "      <th>native_country__Hungary</th>\n", "      <th>native_country__Holand-Netherlands</th>\n", "      <th>native_country__nan</th>\n", "    </tr>\n", "  </thead>\n", "  <tbody>\n", "    <tr>\n", "      <th>10408</th>\n", "      <td>22</td>\n", "      <td>0</td>\n", "      <td>0</td>\n", "      <td>0</td>\n", "      <td>0</td>\n", "      <td>0</td>\n", "      <td>1</td>\n", "      <td>0</td>\n", "      <td>0</td>\n", "      <td>0</td>\n", "      <td>...</td>\n", "      <td>0</td>\n", "      <td>0</td>\n", "      <td>0</td>\n", "      <td>0</td>\n", "      <td>0</td>\n", "      <td>0</td>\n", "      <td>0</td>\n", "      <td>0</td>\n", "      <td>0</td>\n", "      <td>0</td>\n", "    </tr>\n", "    <tr>\n", "      <th>5953</th>\n", "      <td>20</td>\n", "      <td>0</td>\n", "      <td>0</td>\n", "      <td>1</td>\n", "      <td>0</td>\n", "      <td>0</td>\n", "      <td>0</td>\n", "      <td>0</td>\n", "      <td>0</td>\n", "      <td>0</td>\n", "      <td>...</td>\n", "      <td>0</td>\n", "      <td>0</td>\n", "      <td>0</td>\n", "      <td>0</td>\n", "      <td>0</td>\n", "      <td>0</td>\n", "      <td>0</td>\n", "      <td>0</td>\n", "      <td>0</td>\n", "      <td>0</td>\n", "    </tr>\n", "    <tr>\n", "      <th>3059</th>\n", "      <td>22</td>\n", "      <td>0</td>\n", "      <td>0</td>\n", "      <td>1</td>\n", "      <td>0</td>\n", "      <td>0</td>\n", "      <td>0</td>\n", "      <td>0</td>\n", "      <td>0</td>\n", "      <td>0</td>\n", "      <td>...</td>\n", "      <td>0</td>\n", "      <td>0</td>\n", "      <td>0</td>\n", "      <td>0</td>\n", "      <td>0</td>\n", "      <td>0</td>\n", "      <td>0</td>\n", "      <td>0</td>\n", "      <td>0</td>\n", "      <td>0</td>\n", "    </tr>\n", "    <tr>\n", "      <th>11821</th>\n", "      <td>24</td>\n", "      <td>0</td>\n", "      <td>0</td>\n", "      <td>1</td>\n", "      <td>0</td>\n", "      <td>0</td>\n", "      <td>0</td>\n", "      <td>0</td>\n", "      <td>0</td>\n", "      <td>0</td>\n", "      <td>...</td>\n", "      <td>0</td>\n", "      <td>0</td>\n", "      <td>0</td>\n", "      <td>0</td>\n", "      <td>0</td>\n", "      <td>0</td>\n", "      <td>0</td>\n", "      <td>0</td>\n", "      <td>0</td>\n", "      <td>0</td>\n", "    </tr>\n", "    <tr>\n", "      <th>12808</th>\n", "      <td>67</td>\n", "      <td>0</td>\n", "      <td>0</td>\n", "      <td>1</td>\n", "      <td>0</td>\n", "      <td>0</td>\n", "      <td>0</td>\n", "      <td>0</td>\n", "      <td>0</td>\n", "      <td>0</td>\n", "      <td>...</td>\n", "      <td>0</td>\n", "      <td>0</td>\n", "      <td>0</td>\n", "      <td>0</td>\n", "      <td>0</td>\n", "      <td>0</td>\n", "      <td>0</td>\n", "      <td>0</td>\n", "      <td>0</td>\n", "      <td>0</td>\n", "    </tr>\n", "  </tbody>\n", "</table>\n", "<p>5 rows \u00d7 107 columns</p>\n", "</div>"], "text/plain": ["       age  workclass__Self-emp-not-inc  workclass__Private  \\\n", "10408   22                            0                   0   \n", "5953    20                            0                   0   \n", "3059    22                            0                   0   \n", "11821   24                            0                   0   \n", "12808   67                            0                   0   \n", "\n", "       workclass__Federal-gov  workclass__Local-gov  workclass__?  \\\n", "10408                       0                     0             0   \n", "5953                        1                     0             0   \n", "3059                        1                     0             0   \n", "11821                       1                     0             0   \n", "12808                       1                     0             0   \n", "\n", "       workclass__Self-emp-inc  workclass__Without-pay  \\\n", "10408                        1                       0   \n", "5953                         0                       0   \n", "3059                         0                       0   \n", "11821                        0                       0   \n", "12808                        0                       0   \n", "\n", "       workclass__Never-worked  workclass__nan  ...  native_country__Scotland  \\\n", "10408                        0               0  ...                         0   \n", "5953                         0               0  ...                         0   \n", "3059                         0               0  ...                         0   \n", "11821                        0               0  ...                         0   \n", "12808                        0               0  ...                         0   \n", "\n", "       native_country__Trinadad&Tobago  native_country__Greece  \\\n", "10408                                0                       0   \n", "5953                                 0                       0   \n", "3059                                 0                       0   \n", "11821                                0                       0   \n", "12808                                0                       0   \n", "\n", "       native_country__Nicaragua  native_country__Vietnam  \\\n", "10408                          0                        0   \n", "5953                           0                        0   \n", "3059                           0                        0   \n", "11821                          0                        0   \n", "12808                          0                        0   \n", "\n", "       native_country__Hong  native_country__Ireland  native_country__Hungary  \\\n", "10408                     0                        0                        0   \n", "5953                      0                        0                        0   \n", "3059                      0                        0                        0   \n", "11821                     0                        0                        0   \n", "12808                     0                        0                        0   \n", "\n", "       native_country__Holand-Netherlands  native_country__nan  \n", "10408                                   0                    0  \n", "5953                                    0                    0  \n", "3059                                    0                    0  \n", "11821                                   0                    0  \n", "12808                                   0                    0  \n", "\n", "[5 rows x 107 columns]"]}, "execution_count": 53, "metadata": {}, "output_type": "execute_result"}], "source": ["wrong_cat = X_test_cat.iloc[wrong.index, :]\n", "wrong_cat.head()"]}, {"cell_type": "code", "execution_count": 53, "metadata": {}, "outputs": [{"data": {"text/plain": ["array([[2.        , 2.        , 2.23606798, 2.44948974, 2.44948974],\n", "       [1.41421356, 1.41421356, 2.23606798, 2.23606798, 2.23606798],\n", "       [1.        , 1.41421356, 1.41421356, 1.73205081, 1.73205081],\n", "       [1.41421356, 1.73205081, 1.73205081, 1.73205081, 2.        ],\n", "       [2.64575131, 3.31662479, 3.46410162, 3.46410162, 3.74165739]])"]}, "execution_count": 54, "metadata": {}, "output_type": "execute_result"}], "source": ["dist, index = knn.kneighbors(wrong_cat)\n", "dist"]}, {"cell_type": "code", "execution_count": 54, "metadata": {}, "outputs": [{"data": {"text/plain": ["array([[18056, 18362,  8757, 13177,   314],\n", "       [24878, 30060, 27902,  8008, 23771],\n", "       [  920, 31176,  5206,  7415, 15308],\n", "       [17303, 18592,  2019, 20325,  7542],\n", "       [  346, 14421,  8214, 11050, 18902]], dtype=int64)"]}, "execution_count": 55, "metadata": {}, "output_type": "execute_result"}], "source": ["index"]}, {"cell_type": "code", "execution_count": 55, "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>TR-24878</th>\n", "      <th>TR-18056</th>\n", "      <th>TR-920</th>\n", "    </tr>\n", "  </thead>\n", "  <tbody>\n", "    <tr>\n", "      <th>age</th>\n", "      <td>20</td>\n", "      <td>21</td>\n", "      <td>23</td>\n", "    </tr>\n", "    <tr>\n", "      <th>workclass</th>\n", "      <td>Private</td>\n", "      <td>?</td>\n", "      <td>Private</td>\n", "    </tr>\n", "    <tr>\n", "      <th>education</th>\n", "      <td>12th</td>\n", "      <td>Some-college</td>\n", "      <td>Some-college</td>\n", "    </tr>\n", "    <tr>\n", "      <th>education_num</th>\n", "      <td>8</td>\n", "      <td>10</td>\n", "      <td>10</td>\n", "    </tr>\n", "    <tr>\n", "      <th>marital_status</th>\n", "      <td>Never-married</td>\n", "      <td>Never-married</td>\n", "      <td>Never-married</td>\n", "    </tr>\n", "    <tr>\n", "      <th>occupation</th>\n", "      <td>Other-service</td>\n", "      <td>?</td>\n", "      <td>Sales</td>\n", "    </tr>\n", "    <tr>\n", "      <th>relationship</th>\n", "      <td>Own-child</td>\n", "      <td>Own-child</td>\n", "      <td>Not-in-family</td>\n", "    </tr>\n", "    <tr>\n", "      <th>race</th>\n", "      <td>White</td>\n", "      <td>White</td>\n", "      <td>White</td>\n", "    </tr>\n", "    <tr>\n", "      <th>sex</th>\n", "      <td>Male</td>\n", "      <td>Male</td>\n", "      <td>Male</td>\n", "    </tr>\n", "    <tr>\n", "      <th>capital_gain</th>\n", "      <td>0</td>\n", "      <td>0</td>\n", "      <td>0</td>\n", "    </tr>\n", "    <tr>\n", "      <th>capital_loss</th>\n", "      <td>0</td>\n", "      <td>0</td>\n", "      <td>0</td>\n", "    </tr>\n", "    <tr>\n", "      <th>hours_per_week</th>\n", "      <td>35</td>\n", "      <td>16</td>\n", "      <td>25</td>\n", "    </tr>\n", "    <tr>\n", "      <th>native_country</th>\n", "      <td>United-States</td>\n", "      <td>United-States</td>\n", "      <td>United-States</td>\n", "    </tr>\n", "    <tr>\n", "      <th>0</th>\n", "      <td>0</td>\n", "      <td>0</td>\n", "      <td>0</td>\n", "    </tr>\n", "  </tbody>\n", "</table>\n", "</div>"], "text/plain": ["                     TR-24878       TR-18056         TR-920\n", "age                        20             21             23\n", "workclass             Private              ?        Private\n", "education                12th   Some-college   Some-college\n", "education_num               8             10             10\n", "marital_status  Never-married  Never-married  Never-married\n", "occupation      Other-service              ?          Sales\n", "relationship        Own-child      Own-child  Not-in-family\n", "race                    White          White          White\n", "sex                      Male           Male           Male\n", "capital_gain                0              0              0\n", "capital_loss                0              0              0\n", "hours_per_week             35             16             25\n", "native_country  United-States  United-States  United-States\n", "0                           0              0              0"]}, "execution_count": 56, "metadata": {}, "output_type": "execute_result"}], "source": ["train_nn = pandas.concat([X_train, y_train], axis=1).iloc[[24878, 18056, 920], :].T\n", "train_nn.columns = ['TR-' + str(_) for _ in train_nn.columns]\n", "train_nn"]}, {"cell_type": "markdown", "metadata": {}, "source": ["Il faut comparer la premi\u00e8re colonne avec la quatri\u00e8me, la seconde avec la ciinqui\u00e8me et la trois\u00e8me avec la sixi\u00e8me. Ces exemples sont voisins. On voit que les exemples sont tr\u00e8s proches. Il n'y a qu'une seule valeur qui change \u00e0 chaque fois et il est difficile d'expliquer les erreurs."]}, {"cell_type": "code", "execution_count": 56, "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>TR-24878</th>\n", "      <th>TR-18056</th>\n", "      <th>TR-920</th>\n", "      <th>10408</th>\n", "      <th>5953</th>\n", "      <th>3059</th>\n", "    </tr>\n", "  </thead>\n", "  <tbody>\n", "    <tr>\n", "      <th>age</th>\n", "      <td>20</td>\n", "      <td>21</td>\n", "      <td>23</td>\n", "      <td>22</td>\n", "      <td>20</td>\n", "      <td>22</td>\n", "    </tr>\n", "    <tr>\n", "      <th>capital_gain</th>\n", "      <td>0</td>\n", "      <td>0</td>\n", "      <td>0</td>\n", "      <td>0</td>\n", "      <td>0</td>\n", "      <td>0</td>\n", "    </tr>\n", "    <tr>\n", "      <th>capital_loss</th>\n", "      <td>0</td>\n", "      <td>0</td>\n", "      <td>0</td>\n", "      <td>0</td>\n", "      <td>0</td>\n", "      <td>0</td>\n", "    </tr>\n", "    <tr>\n", "      <th>education</th>\n", "      <td>12th</td>\n", "      <td>Some-college</td>\n", "      <td>Some-college</td>\n", "      <td>Some-college</td>\n", "      <td>12th</td>\n", "      <td>Some-college</td>\n", "    </tr>\n", "    <tr>\n", "      <th>education_num</th>\n", "      <td>8</td>\n", "      <td>10</td>\n", "      <td>10</td>\n", "      <td>10</td>\n", "      <td>8</td>\n", "      <td>10</td>\n", "    </tr>\n", "    <tr>\n", "      <th>hours_per_week</th>\n", "      <td>35</td>\n", "      <td>16</td>\n", "      <td>25</td>\n", "      <td>15</td>\n", "      <td>35</td>\n", "      <td>25</td>\n", "    </tr>\n", "    <tr>\n", "      <th>marital_status</th>\n", "      <td>Never-married</td>\n", "      <td>Never-married</td>\n", "      <td>Never-married</td>\n", "      <td>Never-married</td>\n", "      <td>Never-married</td>\n", "      <td>Never-married</td>\n", "    </tr>\n", "    <tr>\n", "      <th>native_country</th>\n", "      <td>United-States</td>\n", "      <td>United-States</td>\n", "      <td>United-States</td>\n", "      <td>?</td>\n", "      <td>United-States</td>\n", "      <td>United-States</td>\n", "    </tr>\n", "    <tr>\n", "      <th>occupation</th>\n", "      <td>Other-service</td>\n", "      <td>?</td>\n", "      <td>Sales</td>\n", "      <td>?</td>\n", "      <td>Other-service</td>\n", "      <td>Sales</td>\n", "    </tr>\n", "    <tr>\n", "      <th>race</th>\n", "      <td>White</td>\n", "      <td>White</td>\n", "      <td>White</td>\n", "      <td>White</td>\n", "      <td>Black</td>\n", "      <td>White</td>\n", "    </tr>\n", "    <tr>\n", "      <th>relationship</th>\n", "      <td>Own-child</td>\n", "      <td>Own-child</td>\n", "      <td>Not-in-family</td>\n", "      <td>Own-child</td>\n", "      <td>Own-child</td>\n", "      <td>Not-in-family</td>\n", "    </tr>\n", "    <tr>\n", "      <th>sex</th>\n", "      <td>Male</td>\n", "      <td>Male</td>\n", "      <td>Male</td>\n", "      <td>Male</td>\n", "      <td>Male</td>\n", "      <td>Male</td>\n", "    </tr>\n", "    <tr>\n", "      <th>workclass</th>\n", "      <td>Private</td>\n", "      <td>?</td>\n", "      <td>Private</td>\n", "      <td>?</td>\n", "      <td>Private</td>\n", "      <td>Private</td>\n", "    </tr>\n", "    <tr>\n", "      <th>0</th>\n", "      <td>0</td>\n", "      <td>0</td>\n", "      <td>0</td>\n", "      <td>NaN</td>\n", "      <td>NaN</td>\n", "      <td>NaN</td>\n", "    </tr>\n", "    <tr>\n", "      <th>y_test</th>\n", "      <td>NaN</td>\n", "      <td>NaN</td>\n", "      <td>NaN</td>\n", "      <td>1</td>\n", "      <td>1</td>\n", "      <td>1</td>\n", "    </tr>\n", "    <tr>\n", "      <th>pred1</th>\n", "      <td>NaN</td>\n", "      <td>NaN</td>\n", "      <td>NaN</td>\n", "      <td>0</td>\n", "      <td>0</td>\n", "      <td>0</td>\n", "    </tr>\n", "    <tr>\n", "      <th>P1(&gt;=50K)</th>\n", "      <td>NaN</td>\n", "      <td>NaN</td>\n", "      <td>NaN</td>\n", "      <td>0.0146811</td>\n", "      <td>0.00972505</td>\n", "      <td>0.0145646</td>\n", "    </tr>\n", "    <tr>\n", "      <th>pred2</th>\n", "      <td>NaN</td>\n", "      <td>NaN</td>\n", "      <td>NaN</td>\n", "      <td>0</td>\n", "      <td>0</td>\n", "      <td>0</td>\n", "    </tr>\n", "    <tr>\n", "      <th>P2(&gt;=50K)</th>\n", "      <td>NaN</td>\n", "      <td>NaN</td>\n", "      <td>NaN</td>\n", "      <td>0</td>\n", "      <td>0</td>\n", "      <td>0</td>\n", "    </tr>\n", "    <tr>\n", "      <th>pred3</th>\n", "      <td>NaN</td>\n", "      <td>NaN</td>\n", "      <td>NaN</td>\n", "      <td>0</td>\n", "      <td>0</td>\n", "      <td>0</td>\n", "    </tr>\n", "    <tr>\n", "      <th>P3(&gt;=50K)</th>\n", "      <td>NaN</td>\n", "      <td>NaN</td>\n", "      <td>NaN</td>\n", "      <td>0.00288558</td>\n", "      <td>0.00299048</td>\n", "      <td>0.00561586</td>\n", "    </tr>\n", "    <tr>\n", "      <th>pred4</th>\n", "      <td>NaN</td>\n", "      <td>NaN</td>\n", "      <td>NaN</td>\n", "      <td>0</td>\n", "      <td>0</td>\n", "      <td>0</td>\n", "    </tr>\n", "    <tr>\n", "      <th>P4(&gt;=50K)</th>\n", "      <td>NaN</td>\n", "      <td>NaN</td>\n", "      <td>NaN</td>\n", "      <td>0.000149386</td>\n", "      <td>0.000395875</td>\n", "      <td>0.000863302</td>\n", "    </tr>\n", "  </tbody>\n", "</table>\n", "</div>"], "text/plain": ["                     TR-24878       TR-18056         TR-920          10408  \\\n", "age                        20             21             23             22   \n", "capital_gain                0              0              0              0   \n", "capital_loss                0              0              0              0   \n", "education                12th   Some-college   Some-college   Some-college   \n", "education_num               8             10             10             10   \n", "hours_per_week             35             16             25             15   \n", "marital_status  Never-married  Never-married  Never-married  Never-married   \n", "native_country  United-States  United-States  United-States              ?   \n", "occupation      Other-service              ?          Sales              ?   \n", "race                    White          White          White          White   \n", "relationship        Own-child      Own-child  Not-in-family      Own-child   \n", "sex                      Male           Male           Male           Male   \n", "workclass             Private              ?        Private              ?   \n", "0                           0              0              0            NaN   \n", "y_test                    NaN            NaN            NaN              1   \n", "pred1                     NaN            NaN            NaN              0   \n", "P1(>=50K)                 NaN            NaN            NaN      0.0146811   \n", "pred2                     NaN            NaN            NaN              0   \n", "P2(>=50K)                 NaN            NaN            NaN              0   \n", "pred3                     NaN            NaN            NaN              0   \n", "P3(>=50K)                 NaN            NaN            NaN     0.00288558   \n", "pred4                     NaN            NaN            NaN              0   \n", "P4(>=50K)                 NaN            NaN            NaN    0.000149386   \n", "\n", "                         5953           3059  \n", "age                        20             22  \n", "capital_gain                0              0  \n", "capital_loss                0              0  \n", "education                12th   Some-college  \n", "education_num               8             10  \n", "hours_per_week             35             25  \n", "marital_status  Never-married  Never-married  \n", "native_country  United-States  United-States  \n", "occupation      Other-service          Sales  \n", "race                    Black          White  \n", "relationship        Own-child  Not-in-family  \n", "sex                      Male           Male  \n", "workclass             Private        Private  \n", "0                         NaN            NaN  \n", "y_test                      1              1  \n", "pred1                       0              0  \n", "P1(>=50K)          0.00972505      0.0145646  \n", "pred2                       0              0  \n", "P2(>=50K)                   0              0  \n", "pred3                       0              0  \n", "P3(>=50K)          0.00299048     0.00561586  \n", "pred4                       0              0  \n", "P4(>=50K)         0.000395875    0.000863302  "]}, "execution_count": 57, "metadata": {}, "output_type": "execute_result"}], "source": ["pandas.concat([train_nn, wrong_study], axis=1, sort=True)"]}, {"cell_type": "markdown", "metadata": {}, "source": ["## Ethique\n", "\n", "Le mod\u00e8le qu'on a appris est-il \u00e9thique ? Il n'y a pas de r\u00e9ponse simples \u00e0 ce probl\u00e8me car il est difficile de transcrire math\u00e9matiquement le caract\u00e8re \u00e9thique d'un mod\u00e8le. Dans le cas pr\u00e9sent, admettons que l'on souhaite v\u00e9rifier que le mod\u00e8le ne retourne pas une r\u00e9ponse biais\u00e9e en fonction du genre. Une premi\u00e8re id\u00e9e consiste \u00e0 enlever la variable pour \u00eatre s\u00fbr sur le mod\u00e8le n'en tienne pas compte mais cela ne garantit que la variable genre n'est une variable corr\u00e9l\u00e9e aux autres. Et la corr\u00e9lation implique ici au sens du mod\u00e8le ce qui a un sens plus fort si le mod\u00e8le n'est pas lin\u00e9aire. On aura tout-\u00e0-faire enlev\u00e9 la variable genre si elle ne peut \u00eatre pr\u00e9dite \u00e0 partir des autres."]}, {"cell_type": "code", "execution_count": 57, "metadata": {}, "outputs": [], "source": ["X_train_sex = train.drop([label, 'sex'], axis=1)\n", "y_train_sex = train['sex'] == 'Male'\n", "X_test_sex = test.drop([label, 'sex'], axis=1)\n", "y_test_sex = test['sex'] == 'Male'"]}, {"cell_type": "code", "execution_count": 58, "metadata": {"scrolled": false}, "outputs": [{"data": {"text/plain": ["Pipeline(memory=None,\n", "         steps=[('onehotencoder',\n", "                 OneHotEncoder(cols=['workclass', 'education', 'marital_status',\n", "                                     'occupation', 'relationship', 'race',\n", "                                     'native_country'],\n", "                               drop_invariant=False, handle_missing='value',\n", "                               handle_unknown='value', return_df=True,\n", "                               use_cat_names=False, verbose=0)),\n", "                ('randomforestclassifier',\n", "                 RandomForestClassifier(bootstrap=True, ccp_alpha=0.0,\n", "                                        class_weight=None, criterion='gini',\n", "                                        max_depth=None, max_features='auto',\n", "                                        max_leaf_nodes=None, max_samples=None,\n", "                                        min_impurity_decrease=0.0,\n", "                                        min_impurity_split=None,\n", "                                        min_samples_leaf=1, min_samples_split=2,\n", "                                        min_weight_fraction_leaf=0.0,\n", "                                        n_estimators=100, n_jobs=None,\n", "                                        oob_score=False, random_state=None,\n", "                                        verbose=0, warm_start=False))],\n", "         verbose=False)"]}, "execution_count": 59, "metadata": {}, "output_type": "execute_result"}], "source": ["ce_sex = OneHotEncoder(cols=[_ for _ in cat_col if _ != 'sex'], \n", "                       handle_missing='value', drop_invariant=False,\n", "                       handle_unknown='value')\n", "model_sex = make_pipeline(ce_sex, RandomForestClassifier(n_estimators=100))\n", "model_sex.fit(X_train_sex, y_train_sex)"]}, {"cell_type": "code", "execution_count": 59, "metadata": {}, "outputs": [{"data": {"text/plain": ["0.8368650574289048"]}, "execution_count": 60, "metadata": {}, "output_type": "execute_result"}], "source": ["model_sex.score(X_test_sex, y_test_sex)"]}, {"cell_type": "markdown", "metadata": {}, "source": ["Il est possible de pr\u00e9dire le genre en fonction des autres variables \u00e0 83% pr\u00e8s. Ce n'est pas en enlevant la variable qu'on peut emp\u00eacher le mod\u00e8le d'\u00eatre biais\u00e9 par rapport \u00e0 cette information. Il n'est pas \u00e9vident de retirer toute influence de ce param\u00e8tre. On peut choisir de la garder en supposant que le mod\u00e8le la choisira plut\u00f4t qu'une autre pour pr\u00e9dire si elle s'av\u00e8re pertinente. Dans ce cas, on pourrait comparer combien de fois le mod\u00e8le change de pr\u00e9diction si on inverse la variable *sex*. On rappelle la performance du mod\u00e8le."]}, {"cell_type": "code", "execution_count": 60, "metadata": {}, "outputs": [{"data": {"text/plain": ["0.8448498249493275"]}, "execution_count": 61, "metadata": {}, "output_type": "execute_result"}], "source": ["pipe2.score(X_test, y_test)"]}, {"cell_type": "markdown", "metadata": {}, "source": ["On remplace la variable *sex* par la pr\u00e9diction de l'autre mod\u00e8le."]}, {"cell_type": "code", "execution_count": 61, "metadata": {}, "outputs": [], "source": ["X_test_modified = X_test.copy()\n", "X_test_modified['sex'] = model_sex.predict(X_test_sex)\n", "X_test_modified['sex'] = X_test_modified['sex'].apply(lambda x: 'Male' if x else 'Female')"]}, {"cell_type": "code", "execution_count": 62, "metadata": {}, "outputs": [{"data": {"text/plain": ["0.843928505620048"]}, "execution_count": 63, "metadata": {}, "output_type": "execute_result"}], "source": ["pipe2.score(X_test_modified, y_test)"]}, {"cell_type": "markdown", "metadata": {}, "source": ["Quasiment aucun changement. Inversons la colonne maintenant."]}, {"cell_type": "code", "execution_count": 63, "metadata": {}, "outputs": [], "source": ["X_test_inv = X_test.copy()\n", "X_test_inv['sex'] = X_test_inv['sex'].apply(lambda x: 'Female' if x == 'Male' else 'Male')"]}, {"cell_type": "code", "execution_count": 64, "metadata": {}, "outputs": [{"data": {"text/plain": ["0.8505005835022419"]}, "execution_count": 65, "metadata": {}, "output_type": "execute_result"}], "source": ["pipe2.score(X_test_inv, y_test)"]}, {"cell_type": "markdown", "metadata": {}, "source": ["Encore mieux. Regardons les diff\u00e9rences maintenant."]}, {"cell_type": "code", "execution_count": 65, "metadata": {}, "outputs": [{"data": {"text/plain": ["(1024, 16281, 0.06289539954548247)"]}, "execution_count": 66, "metadata": {}, "output_type": "execute_result"}], "source": ["diff1 = X_test['sex'] != X_test_inv['sex']\n", "diff2 = pipe2.predict(X_test) != pipe2.predict(X_test_inv)\n", "diff2.sum(), diff1.sum(), diff2.sum() / diff1.sum()"]}, {"cell_type": "markdown", "metadata": {}, "source": ["La pr\u00e9diction change dans 5% des cas. On est s\u00fbr que pour ces observations et ce mod\u00e8le, le genre a un impact, cela ne veut rien dire pour les autres. Regardons les cinq premi\u00e8res."]}, {"cell_type": "code", "execution_count": 66, "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>2</th>\n", "      <th>14</th>\n", "      <th>17</th>\n", "      <th>19</th>\n", "      <th>28</th>\n", "    </tr>\n", "  </thead>\n", "  <tbody>\n", "    <tr>\n", "      <th>age</th>\n", "      <td>28</td>\n", "      <td>48</td>\n", "      <td>43</td>\n", "      <td>40</td>\n", "      <td>54</td>\n", "    </tr>\n", "    <tr>\n", "      <th>workclass</th>\n", "      <td>Local-gov</td>\n", "      <td>Private</td>\n", "      <td>Private</td>\n", "      <td>Private</td>\n", "      <td>Private</td>\n", "    </tr>\n", "    <tr>\n", "      <th>education</th>\n", "      <td>Assoc-acdm</td>\n", "      <td>HS-grad</td>\n", "      <td>HS-grad</td>\n", "      <td>Doctorate</td>\n", "      <td>HS-grad</td>\n", "    </tr>\n", "    <tr>\n", "      <th>education_num</th>\n", "      <td>12</td>\n", "      <td>9</td>\n", "      <td>9</td>\n", "      <td>16</td>\n", "      <td>9</td>\n", "    </tr>\n", "    <tr>\n", "      <th>marital_status</th>\n", "      <td>Married-civ-spouse</td>\n", "      <td>Married-civ-spouse</td>\n", "      <td>Married-civ-spouse</td>\n", "      <td>Married-civ-spouse</td>\n", "      <td>Married-civ-spouse</td>\n", "    </tr>\n", "    <tr>\n", "      <th>occupation</th>\n", "      <td>Protective-serv</td>\n", "      <td>Machine-op-inspct</td>\n", "      <td>Adm-clerical</td>\n", "      <td>Prof-specialty</td>\n", "      <td>Craft-repair</td>\n", "    </tr>\n", "    <tr>\n", "      <th>relationship</th>\n", "      <td>Husband</td>\n", "      <td>Husband</td>\n", "      <td>Wife</td>\n", "      <td>Husband</td>\n", "      <td>Husband</td>\n", "    </tr>\n", "    <tr>\n", "      <th>race</th>\n", "      <td>White</td>\n", "      <td>White</td>\n", "      <td>White</td>\n", "      <td>Asian-Pac-Islander</td>\n", "      <td>White</td>\n", "    </tr>\n", "    <tr>\n", "      <th>sex</th>\n", "      <td>Male</td>\n", "      <td>Male</td>\n", "      <td>Female</td>\n", "      <td>Male</td>\n", "      <td>Male</td>\n", "    </tr>\n", "    <tr>\n", "      <th>capital_gain</th>\n", "      <td>0</td>\n", "      <td>3103</td>\n", "      <td>0</td>\n", "      <td>0</td>\n", "      <td>0</td>\n", "    </tr>\n", "    <tr>\n", "      <th>capital_loss</th>\n", "      <td>0</td>\n", "      <td>0</td>\n", "      <td>0</td>\n", "      <td>0</td>\n", "      <td>0</td>\n", "    </tr>\n", "    <tr>\n", "      <th>hours_per_week</th>\n", "      <td>40</td>\n", "      <td>48</td>\n", "      <td>30</td>\n", "      <td>45</td>\n", "      <td>35</td>\n", "    </tr>\n", "    <tr>\n", "      <th>native_country</th>\n", "      <td>United-States</td>\n", "      <td>United-States</td>\n", "      <td>United-States</td>\n", "      <td>?</td>\n", "      <td>United-States</td>\n", "    </tr>\n", "    <tr>\n", "      <th>y</th>\n", "      <td>1</td>\n", "      <td>1</td>\n", "      <td>0</td>\n", "      <td>1</td>\n", "      <td>0</td>\n", "    </tr>\n", "    <tr>\n", "      <th>prediction_sex</th>\n", "      <td>True</td>\n", "      <td>True</td>\n", "      <td>False</td>\n", "      <td>True</td>\n", "      <td>True</td>\n", "    </tr>\n", "  </tbody>\n", "</table>\n", "</div>"], "text/plain": ["                                2                   14                  17  \\\n", "age                             28                  48                  43   \n", "workclass                Local-gov             Private             Private   \n", "education               Assoc-acdm             HS-grad             HS-grad   \n", "education_num                   12                   9                   9   \n", "marital_status  Married-civ-spouse  Married-civ-spouse  Married-civ-spouse   \n", "occupation         Protective-serv   Machine-op-inspct        Adm-clerical   \n", "relationship               Husband             Husband                Wife   \n", "race                         White               White               White   \n", "sex                           Male                Male              Female   \n", "capital_gain                     0                3103                   0   \n", "capital_loss                     0                   0                   0   \n", "hours_per_week                  40                  48                  30   \n", "native_country       United-States       United-States       United-States   \n", "y                                1                   1                   0   \n", "prediction_sex                True                True               False   \n", "\n", "                                19                  28  \n", "age                             40                  54  \n", "workclass                  Private             Private  \n", "education                Doctorate             HS-grad  \n", "education_num                   16                   9  \n", "marital_status  Married-civ-spouse  Married-civ-spouse  \n", "occupation          Prof-specialty        Craft-repair  \n", "relationship               Husband             Husband  \n", "race            Asian-Pac-Islander               White  \n", "sex                           Male                Male  \n", "capital_gain                     0                   0  \n", "capital_loss                     0                   0  \n", "hours_per_week                  45                  35  \n", "native_country                   ?       United-States  \n", "y                                1                   0  \n", "prediction_sex                True                True  "]}, "execution_count": 67, "metadata": {}, "output_type": "execute_result"}], "source": ["look = X_test.copy()\n", "look['y'] = y_test\n", "look['prediction_sex'] = model_sex.predict(X_test_sex)\n", "look[diff2].head().T"]}, {"cell_type": "markdown", "metadata": {}, "source": ["Il existe visible quelques observations \u00e0 v\u00e9rifier o\u00f9 la relation *relationship* et le genre *sex* semble \u00eatre en contradiction ou plut\u00f4t ne pas prendre en compte tous les types de relations possibles."]}, {"cell_type": "code", "execution_count": 67, "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>relationship</th>\n", "      <th>Husband</th>\n", "      <th>Not-in-family</th>\n", "      <th>Other-relative</th>\n", "      <th>Own-child</th>\n", "      <th>Unmarried</th>\n", "      <th>Wife</th>\n", "    </tr>\n", "    <tr>\n", "      <th>sex</th>\n", "      <th></th>\n", "      <th></th>\n", "      <th></th>\n", "      <th></th>\n", "      <th></th>\n", "      <th></th>\n", "    </tr>\n", "  </thead>\n", "  <tbody>\n", "    <tr>\n", "      <th>Female</th>\n", "      <td>1</td>\n", "      <td>3875</td>\n", "      <td>430</td>\n", "      <td>2245</td>\n", "      <td>2654</td>\n", "      <td>1566</td>\n", "    </tr>\n", "    <tr>\n", "      <th>Male</th>\n", "      <td>13192</td>\n", "      <td>4430</td>\n", "      <td>551</td>\n", "      <td>2823</td>\n", "      <td>792</td>\n", "      <td>2</td>\n", "    </tr>\n", "  </tbody>\n", "</table>\n", "</div>"], "text/plain": ["relationship  Husband  Not-in-family  Other-relative  Own-child  Unmarried  \\\n", "sex                                                                          \n", "Female              1           3875             430       2245       2654   \n", "Male            13192           4430             551       2823        792   \n", "\n", "relationship  Wife  \n", "sex                 \n", "Female        1566  \n", "Male             2  "]}, "execution_count": 68, "metadata": {}, "output_type": "execute_result"}], "source": ["X_train[['sex', 'relationship', 'age']].groupby(['sex', 'relationship'], as_index=False)\\\n", "        .count().pivot('sex', 'relationship', 'age')"]}, {"cell_type": "markdown", "metadata": {}, "source": ["Pas de conclusion \u00e0 ce stade. Il faut poursuivre l'exploration [Machine Learning \u00e9thique](http://www.xavierdupre.fr/app/ensae_teaching_cs/helpsphinx/td_2a_mlplus.html#machine-learning-ethique)."]}, {"cell_type": "markdown", "metadata": {}, "source": ["## S\u00e9lection des variables\n", "\n", "Il n'y a pas de m\u00e9thode optimale pour s\u00e9lectionner les variables. Il existe diff\u00e9rentes options comme celles propos\u00e9es par [scikit-learn/feature_selection](http://scikit-learn.org/stable/modules/classes.html#module-sklearn.feature_selection). Certaines partent des features brutes, d'autres utilisent le mod\u00e8le qui doit \u00eatre appris. Mais ce n'est pas toujours \u00e9vident de faire marcher ces m\u00e9thodes sur n'importe quel mod\u00e8le."]}, {"cell_type": "code", "execution_count": 68, "metadata": {}, "outputs": [{"name": "stdout", "output_type": "stream", "text": ["could not convert string to float: 'State-gov'\n"]}], "source": ["from sklearn.feature_selection import RFE\n", "try:\n", "    model = RFE(pipe2)\n", "    model.fit(X_train, y_train)\n", "except Exception as e:\n", "    print(e)"]}, {"cell_type": "markdown", "metadata": {}, "source": ["Dans notre cas, on retire les variables une \u00e0 une en fonction de l'indicateur ``feature_importance`` ce qui n'est pas facile car les variables sont des modalit\u00e9s. Il faut en faire la somme..."]}, {"cell_type": "code", "execution_count": 69, "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>raw_var</th>\n", "      <th>importance</th>\n", "    </tr>\n", "  </thead>\n", "  <tbody>\n", "    <tr>\n", "      <th>0</th>\n", "      <td>race</td>\n", "      <td>0.018372</td>\n", "    </tr>\n", "    <tr>\n", "      <th>1</th>\n", "      <td>sex</td>\n", "      <td>0.019006</td>\n", "    </tr>\n", "    <tr>\n", "      <th>2</th>\n", "      <td>native_country</td>\n", "      <td>0.028188</td>\n", "    </tr>\n", "    <tr>\n", "      <th>3</th>\n", "      <td>capital_loss</td>\n", "      <td>0.034305</td>\n", "    </tr>\n", "    <tr>\n", "      <th>4</th>\n", "      <td>workclass</td>\n", "      <td>0.047340</td>\n", "    </tr>\n", "    <tr>\n", "      <th>5</th>\n", "      <td>education</td>\n", "      <td>0.059821</td>\n", "    </tr>\n", "    <tr>\n", "      <th>6</th>\n", "      <td>education_num</td>\n", "      <td>0.064527</td>\n", "    </tr>\n", "    <tr>\n", "      <th>7</th>\n", "      <td>relationship</td>\n", "      <td>0.083855</td>\n", "    </tr>\n", "    <tr>\n", "      <th>8</th>\n", "      <td>occupation</td>\n", "      <td>0.091443</td>\n", "    </tr>\n", "    <tr>\n", "      <th>9</th>\n", "      <td>marital_status</td>\n", "      <td>0.100458</td>\n", "    </tr>\n", "    <tr>\n", "      <th>10</th>\n", "      <td>capital_gain</td>\n", "      <td>0.110967</td>\n", "    </tr>\n", "    <tr>\n", "      <th>11</th>\n", "      <td>hours_per_week</td>\n", "      <td>0.113923</td>\n", "    </tr>\n", "    <tr>\n", "      <th>12</th>\n", "      <td>age</td>\n", "      <td>0.227797</td>\n", "    </tr>\n", "  </tbody>\n", "</table>\n", "</div>"], "text/plain": ["           raw_var  importance\n", "0             race    0.018372\n", "1              sex    0.019006\n", "2   native_country    0.028188\n", "3     capital_loss    0.034305\n", "4        workclass    0.047340\n", "5        education    0.059821\n", "6    education_num    0.064527\n", "7     relationship    0.083855\n", "8       occupation    0.091443\n", "9   marital_status    0.100458\n", "10    capital_gain    0.110967\n", "11  hours_per_week    0.113923\n", "12             age    0.227797"]}, "execution_count": 70, "metadata": {}, "output_type": "execute_result"}], "source": ["def grouped_feature_importance(model, datas, cat_col):\n", "    ce = model.steps[0][-1]\n", "    data_cat = ce.fit_transform(datas)    \n", "    rename_columns(data_cat, ce)\n", "    df = pandas.DataFrame(dict(name=data_cat.columns, \n", "                               importance=model.steps[-1][-1].feature_importances_))\n", "    df = df.sort_values(\"importance\", ascending=False).reset_index(drop=True)\n", "    df['raw_var'] = df['name'].apply(lambda x: x.split('__')[0])\n", "    gr = df.groupby('raw_var').sum().sort_values('importance', ascending=True).reset_index(drop=False).copy()\n", "    return gr\n", "\n", "fi_global = grouped_feature_importance(pipe2, X_train, cat_col)\n", "fi_global"]}, {"cell_type": "markdown", "metadata": {}, "source": ["On fait une boucle."]}, {"cell_type": "code", "execution_count": 70, "metadata": {}, "outputs": [{"name": "stdout", "output_type": "stream", "text": ["13 0.8345310484613967 None ['race', 0.017876461431527595] (32561, 13)\n", "12 0.8328112523800749 race ['native_country', 0.030328521242080998] (32561, 12)\n", "11 0.8377863767581843 native_country ['sex', 0.020085260379222935] (32561, 11)\n", "10 0.8366193722744303 sex ['capital_loss', 0.04176572757064676] (32561, 10)\n", "9 0.8286960260426264 capital_loss ['workclass', 0.051745962436835796] (32561, 9)\n", "8 0.8289417111971009 workclass ['education', 0.06708986825516294] (32561, 8)\n", "7 0.8292488176401941 education ['marital_status', 0.10470565098684484] (32561, 7)\n", "6 0.8299858731036177 marital_status ['education_num', 0.12482378792123464] (32561, 6)\n", "5 0.8271604938271605 education_num ['capital_gain', 0.15606710618140648] (32561, 5)\n", "4 0.8043731957496468 capital_gain ['hours_per_week', 0.1952915897722717] (32561, 4)\n", "3 0.8129721761562557 hours_per_week ['age', 0.2703491413657654] (32561, 3)\n", "2 0.8207726798108225 age ['occupation', 0.3492816128792312] (32561, 2)\n", "1 0.7637737239727289 occupation ['relationship', 0.9999999999999999] (32561, 1)\n"]}], "source": ["kept = list(fi_global.raw_var)\n", "res = []\n", "last_removed = None\n", "while len(kept) > 0:\n", "    cat_col_red = set()\n", "    for col in kept:\n", "        if \"__\" in col:\n", "            col = \"__\".join(col.split('__')[:-1])\n", "        cat_col_red.add(col)\n", "    cat_col_red = list(cat_col_red)\n", "    X_train_reduced = X_train[cat_col_red]\n", "    X_test_reduced = X_test[cat_col_red]    \n", "    ce = OneHotEncoder(cols=cat_col_red, handle_missing='value',\n", "                       drop_invariant=False, handle_unknown='value')    \n", "    model = make_pipeline(ce, RandomForestClassifier(n_estimators=5))\n", "    model.fit(X_train_reduced, y_train)\n", "    score = model.score(X_test_reduced, y_test)\n", "    fi = grouped_feature_importance(model, X_train_reduced, cat_col_red)\n", "    r = dict(score=score, features=kept.copy(), nb=len(kept), model=model,\n", "             removed=last_removed, next_remove=fi.iloc[0,0], score_remove=fi.iloc[0,1])\n", "    print(r['nb'], r['score'], last_removed, list(fi.iloc[0,:]), X_train_reduced.shape)\n", "    last_removed = fi.iloc[0,0]\n", "    kept = [_ for _ in kept if _ != last_removed]\n", "    res.append(r)"]}, {"cell_type": "markdown", "metadata": {}, "source": ["Il faut se rappeler qu'un classifieur constant retournerait environ 76% de bonne classification."]}, {"cell_type": "code", "execution_count": 71, "metadata": {}, "outputs": [{"data": {"text/plain": ["0.7637737239727289"]}, "execution_count": 72, "metadata": {}, "output_type": "execute_result"}], "source": ["1 - y_test.sum() / len(y_test)"]}, {"cell_type": "code", "execution_count": 72, "metadata": {}, "outputs": [], "source": []}], "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": "3.7.2"}}, "nbformat": 4, "nbformat_minor": 2}