{
 "cells": [
  {
   "cell_type": "markdown",
   "id": "c1c6b6b0-267f-49b3-b70c-da5fbda471a4",
   "metadata": {},
   "source": [
    "# Analyse de points d'intérêts (POI) : Autoencoder, espace latent et embending"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "92f23e03-2645-47be-bafe-5cf410375668",
   "metadata": {},
   "source": [
    "Le notebook : https://sergelhomme.fr/notebook/Intro_IA_poi.ipynb"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "df137542-df2f-4f50-9929-200ed83e23a9",
   "metadata": {},
   "source": [
    "En géographie et en aménagement, les principaux exemples mobilisés pour illustrer l'intérêt des réseaux de neurones portent généralement sur le traitement d'images, notamment en télédétection. En matière d'enseignement, les modèles de prédiction économique constituent également des supports pédagogiques intéressants pour appréhender le fonctionnement des réseaux de neurones et développer ses premiers modèles. Néanmoins, depuis la fin des années 2010, l'adaptation de modèles de plongement lexical tels que Word2Vec aux données spatiales a favorisé l'émergence d'un champ de recherche original, encore relativement peu mis en avant : l'analyse spatiale des points d'intérêt (POI). L'objectif est alors de caractériser (de classer) les territoires à partir de ces POIs ou de mieux comprendre la logique d'implantation de ces activités. Dit de manière érudite, il s'agit de saisir la sémantique de l'espace ! On cherche à définir le « sens » d'un lieu — sa fonction urbaine — en analysant les types de commerces, de services ou d'infrastructures qui l'entourent et en traduisant la configuration spatiale d'un territoire en « concepts sémantiques »."
   ]
  },
  {
   "cell_type": "markdown",
   "id": "e7e3b896-c48a-4e83-8586-d7f889ee673e",
   "metadata": {},
   "source": [
    "Les recherches récentes sur les POI s'intéressent notamment à la classification des espaces géographiques. Cette problématique n'est toutefois pas nouvelle en géographie, où les méthodes de statistique multivariée diffusées lors de la révolution quantitative des années 1960 occupent une place importante. Les analyses en composantes principales (ACP), parfois associées à des classifications ascendantes hiérarchiques (CAH) ou à des algorithmes de partitionnement tels que le k-means, constituent ainsi des outils classiques de typologie territoriale. Leur principal intérêt réside dans leur caractère exploratoire : les regroupements sont déterminés à partir des régularités observées dans les données plutôt qu'à partir d'une classification définie a priori. Ces approches présentent néanmoins plusieurs limites. Les résultats dépendent fortement des choix méthodologiques effectués (sélection et transformation des variables, métrique de distance, nombre de classes retenues), tandis que certaines méthodes reposent sur des hypothèses de linéarité qui peuvent s'avérer restrictives face à la complexité des phénomènes spatiaux."
   ]
  },
  {
   "cell_type": "markdown",
   "id": "567a38f2-f092-49f5-9c48-34f6e9be3bd3",
   "metadata": {},
   "source": [
    "Cette dernière limite est particulièrement importante dans la mesure où la géographie s'intéresse à des phénomènes spatiaux complexes dont les relations ne sont pas nécessairement linéaires. Dans cette perspective, les réseaux de neurones apparaissent comme une alternative intéressante grâce à leur capacité à apprendre des représentations non linéaires des données. Toutefois, ces méthodes sont souvent associées à l'apprentissage supervisé et aux tâches de prédiction nécessitant des données préalablement étiquetées. Cette caractéristique semble a priori peu compatible avec l'objectif exploratoire des classifications territoriales, puisqu'elle implique de fournir au modèle des catégories préexistantes servant de référence pour l'apprentissage (ce qui serait possible, car on peut imaginer demander à des experts de classer une partie des lieux d'un territoire, puis l'objectif du réseau sera d'apprendre la logique des experts et de réaliser la classification à grande échelle). Or, l'un des intérêts des démarches de classification en géographie est précisément de faire émerger des regroupements susceptibles de questionner ces catégories expertes ou institutionnelles."
   ]
  },
  {
   "cell_type": "markdown",
   "id": "ab65182c-a820-4cec-b523-8969d1e2d9ca",
   "metadata": {},
   "source": [
    "## Autoencoder et espace latent : classification non supervisée non linéaire"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "e92e75d5-7e5a-48a9-8172-132cdbd16c80",
   "metadata": {},
   "source": [
    "Les travaux menés dans le domaine de l'apprentissage profond ont néanmoins permis de dépasser en partie cette opposition. Parmi les avancées les plus importantes figurent les autoencodeurs, notamment développés et popularisés par Geoffrey Hinton et ses collaborateurs, qui permettent d'apprendre automatiquement des représentations latentes des données sans recourir à des labels. Pour cela, les autoencodeurs vont chercher à transformer les données de telle manière à les simplifier au maximum puis à reconstruire ces données pour retrouver le résultat initial. Un autoencodeur est un réseau de neurones qui prend un objet complexe, le contraint à passer dans une sorte d'entonnoir informationnel, puis tente d'en reconstruire une version aussi fidèle que possible. Il apprend ainsi à partir des données elles-mêmes, en cherchant à reconstruire son entrée. La première partie du réseau, appelée encodeur, est constituée de couches de neurones de taille décroissante jusqu'à atteindre la couche la plus petite, appelée espace latent ou couche latente. À ce stade, l'information a été compressée sous une forme plus compacte. La seconde partie, appelée décodeur, est alors chargée de reconstruire l'objet initial à partir de cette représentation condensée grâce à une succession de couches de neurones de taille croissante."
   ]
  },
  {
   "cell_type": "markdown",
   "id": "52fae900-8010-4526-806d-b355698a6167",
   "metadata": {},
   "source": [
    "<b> Pour faire très simple, un réseau de neurones doit être guidé pour apprendre, il doit être supervisé. Il doit pouvoir faire des prédictions dont il peut vérifier la justesse (X -> Y). Si on ne peut pas lui donner en entrée des labels (des Y, des données étiquetées, des prédictions à faire), les autoencodeurs sont une solution. Ils vont alors simplifier (compresser) les données qu'on leur donne, puis ils vont chercher à reconstruire les données à partir de cette simplification pour vérifier s'ils ont bien simplifié, s'ils ont bien compris (X -> Compression (Espace Latent) -> X). On utilisera l'espace latent pour produire Y (une prédiction qui n'en est pas vraiment une, une simple transformation).</b>"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "12f7601b-a5bb-4568-a9ee-1ec43e6ce437",
   "metadata": {},
   "source": [
    "Bien que le principe puisse paraître complexe, sa mise en œuvre est aujourd'hui relativement simple grâce aux bibliothèques disponibles en Python. En entrée, l'autoencodeur utilise le même type de tableau de données que celui mobilisé dans une classification non supervisée fondée sur une ACP. Les lignes correspondent aux objets géographiques que l'on souhaite caractériser et les colonnes aux variables descriptives. Dans cet exemple, nous nous appuyons sur les données de la Base Permanente des Équipements (BPE) pour l'Île-de-France. Une jointure spatiale est d'abord réalisée entre les équipements et les communes franciliennes à l'aide de la fonction sjoin(). La fonction crosstab() permet ensuite de construire un tableau de dénombrement croisant les communes et les types d'équipements. Les lignes correspondent alors aux communes, les colonnes aux catégories d'équipements, et chaque cellule indique le nombre d'équipements d'un type donné présents dans une commune. Les données sont téléchargeables <a href='https://sergelhomme.fr/data/IA.zip'>ici</a>."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "id": "804c45ab-b01b-4c82-a3b2-b9b1070127b0",
   "metadata": {},
   "outputs": [],
   "source": [
    "import geopandas as gpd\n",
    "import pandas as pd\n",
    "import numpy as np\n",
    "\n",
    "zones = gpd.read_file(\"COMMUNE-IDF.shp\") \n",
    "poi = gpd.read_file(\"BPE_IDF.shp\") \n",
    "\n",
    "poi_dans_zones = gpd.sjoin(poi, zones, how=\"inner\", predicate=\"within\")\n",
    "matrice_brute = pd.crosstab(poi_dans_zones['INSEE_COM'], poi_dans_zones['Type']) "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "id": "aeef2c28-f2a5-4ede-a246-835f5fcddfe2",
   "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>Type</th>\n",
       "      <th>A101</th>\n",
       "      <th>A104</th>\n",
       "      <th>A105</th>\n",
       "      <th>A108</th>\n",
       "      <th>A109</th>\n",
       "      <th>A120</th>\n",
       "      <th>A121</th>\n",
       "      <th>A122</th>\n",
       "      <th>A124</th>\n",
       "      <th>A125</th>\n",
       "      <th>...</th>\n",
       "      <th>F307</th>\n",
       "      <th>F312</th>\n",
       "      <th>F313</th>\n",
       "      <th>F314</th>\n",
       "      <th>F315</th>\n",
       "      <th>G101</th>\n",
       "      <th>G102</th>\n",
       "      <th>G103</th>\n",
       "      <th>G104</th>\n",
       "      <th>G105</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>INSEE_COM</th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></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>75101</th>\n",
       "      <td>9</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>0</td>\n",
       "      <td>...</td>\n",
       "      <td>1</td>\n",
       "      <td>4</td>\n",
       "      <td>4</td>\n",
       "      <td>0</td>\n",
       "      <td>7</td>\n",
       "      <td>78</td>\n",
       "      <td>81</td>\n",
       "      <td>0</td>\n",
       "      <td>44</td>\n",
       "      <td>3</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>75102</th>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>4</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>...</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>5</td>\n",
       "      <td>62</td>\n",
       "      <td>37</td>\n",
       "      <td>0</td>\n",
       "      <td>31</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>75103</th>\n",
       "      <td>2</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>0</td>\n",
       "      <td>...</td>\n",
       "      <td>2</td>\n",
       "      <td>5</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>2</td>\n",
       "      <td>34</td>\n",
       "      <td>31</td>\n",
       "      <td>0</td>\n",
       "      <td>34</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>75104</th>\n",
       "      <td>5</td>\n",
       "      <td>0</td>\n",
       "      <td>1</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>...</td>\n",
       "      <td>4</td>\n",
       "      <td>6</td>\n",
       "      <td>4</td>\n",
       "      <td>0</td>\n",
       "      <td>6</td>\n",
       "      <td>19</td>\n",
       "      <td>40</td>\n",
       "      <td>0</td>\n",
       "      <td>32</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>75105</th>\n",
       "      <td>3</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>0</td>\n",
       "      <td>...</td>\n",
       "      <td>4</td>\n",
       "      <td>8</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>6</td>\n",
       "      <td>36</td>\n",
       "      <td>80</td>\n",
       "      <td>0</td>\n",
       "      <td>32</td>\n",
       "      <td>2</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>5 rows × 228 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "Type       A101  A104  A105  A108  A109  A120  A121  A122  A124  A125  ...  \\\n",
       "INSEE_COM                                                              ...   \n",
       "75101         9     0     0     0     0     1     0     0     0     0  ...   \n",
       "75102         1     0     0     0     0     4     0     1     0     0  ...   \n",
       "75103         2     0     0     0     0     1     0     0     0     0  ...   \n",
       "75104         5     0     1     0     1     0     0     0     0     0  ...   \n",
       "75105         3     0     0     0     0     1     0     0     0     0  ...   \n",
       "\n",
       "Type       F307  F312  F313  F314  F315  G101  G102  G103  G104  G105  \n",
       "INSEE_COM                                                              \n",
       "75101         1     4     4     0     7    78    81     0    44     3  \n",
       "75102         1     0     0     0     5    62    37     0    31     0  \n",
       "75103         2     5     0     1     2    34    31     0    34     0  \n",
       "75104         4     6     4     0     6    19    40     0    32     1  \n",
       "75105         4     8     1     0     6    36    80     0    32     2  \n",
       "\n",
       "[5 rows x 228 columns]"
      ]
     },
     "execution_count": 2,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "matrice_brute.head()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "b315a459-19ee-4dc5-8b81-1b10635df9d4",
   "metadata": {},
   "source": [
    "Comme dans toute démarche de classification statistique, plusieurs questions se posent concernant le traitement des données : faut-il utiliser les données brutes ou les transformer ? Une normalisation est-elle nécessaire et, si oui, laquelle retenir ? Les réseaux de neurones, bien qu'ils permettent d'apprendre des relations non linéaires, ne dispensent pas de ces choix méthodologiques. Les résultats obtenus demeurent fortement dépendants de la manière dont les données sont préparées. Dans notre cas, il paraît pertinent de travailler sur des données relatives obtenues par une normalisation ligne par ligne. Sans cette étape, le réseau aurait tendance à regrouper les communes principalement en fonction de leur nombre total d'équipements plutôt qu'en fonction de la structure de leur offre d'équipements. Toutefois, une normalisation par ligne présente également un inconvénient : elle conduit à considérer comme identiques deux communes présentant la même répartition d'équipements, même si l'une en possède dix fois plus d'équipements que l'autre. Afin de conserver cette information de taille, nous ajoutons donc une variable supplémentaire correspondant au nombre total d'équipements présents dans chaque commune."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "id": "46860c5e-aa01-4fbb-a25f-242078c71179",
   "metadata": {},
   "outputs": [],
   "source": [
    "from sklearn.preprocessing import Normalizer\n",
    "scaler = Normalizer(norm='l1') \n",
    "X = scaler.fit_transform(matrice_brute)\n",
    "taille = matrice_brute.sum(axis=1).to_numpy() # Prise en compte de la taille\n",
    "X = np.column_stack((X, taille))\n",
    "nb_variables = X.shape[1] "
   ]
  },
  {
   "cell_type": "markdown",
   "id": "d37d6095-5983-4204-a740-5720fcaafaf5",
   "metadata": {},
   "source": [
    "Nous disposons ainsi de 229 variables descriptives, que nous choisissons de projeter dans un espace latent de cinq dimensions. Pourquoi cinq dimensions ? Nous reviendrons en partie sur cette question lors de l'analyse des résultats. Pour réaliser cette compression, l'encodeur comporte une couche intermédiaire de 32 neurones. D'autres architectures auraient pu être envisagées, avec 16, 64 ou 128 neurones, voire plusieurs couches intermédiaires successives. En pratique, ces choix relèvent largement d'une démarche empirique et sont guidés par la qualité des résultats obtenus, notamment en termes de reconstruction et d'interprétation. La fonction d'activation retenue est une fonction ReLU, couramment utilisée dans ce type d'application. L'encodeur étant défini, le décodeur est construit de manière symétrique afin de reconstruire les données initiales à partir de l'espace latent. Seule la couche de sortie diffère et utilise une activation linéaire, adaptée à la reconstruction de variables quantitatives continues. Si le jeu de données n'avait contenu que les variables normalisées comprises entre 0 et 1, une fonction sigmoïde aurait également pu être envisagée pour la couche de sortie."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "id": "4817adc3-9dee-46c2-88f0-881160902e98",
   "metadata": {},
   "outputs": [],
   "source": [
    "import tensorflow as tf\n",
    "from tensorflow.keras import layers, models\n",
    "\n",
    "dimension_latente = 5\n",
    "\n",
    "input_layer = layers.Input(shape=(nb_variables,))\n",
    "# Encodeur : Compresse les données via des fonctions non linéaires (ReLU)\n",
    "encoded = layers.Dense(32, activation='relu')(input_layer)\n",
    "bottleneck = layers.Dense(dimension_latente, activation='relu', name=\"couche_latente\")(encoded)\n",
    "# Décodeur : Tente de reconstruire le vecteur d'origine à partir de la compression\n",
    "decoded = layers.Dense(32, activation='relu')(bottleneck)\n",
    "output_layer = layers.Dense(nb_variables, activation='linear')(decoded)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "9bd36d75-d919-41db-9001-45745fd6dac6",
   "metadata": {},
   "source": [
    "Il convient également de noter que le réseau a été implémenté à l'aide de l'API fonctionnelle de Keras plutôt qu'avec l'API séquentielle. Dans le cas présent, l'autoencodeur aurait pu être défini sous forme séquentielle sans difficulté particulière. Toutefois, l'API fonctionnelle offre davantage de souplesse pour construire des architectures plus complexes, notamment lorsque plusieurs couches, entrées ou sorties doivent être combinées. Une fois le réseau défini, le modèle est compilé puis entraîné de manière classique en minimisant l'erreur de reconstruction entre les données d'entrée et les données reconstruites par le décodeur."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "id": "76f2848b-462f-4766-8c87-be76f7d05c69",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\u001b[1m41/41\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 2ms/step\n"
     ]
    }
   ],
   "source": [
    "# Assemblage et compilation du modèle complet\n",
    "autoencoder = models.Model(inputs=input_layer, outputs=output_layer)\n",
    "autoencoder.compile(optimizer='adam', loss='mse')\n",
    "# Entraînement : Le modèle apprend à prédire X à partir de X (auto-supervision)\n",
    "history = autoencoder.fit(X, X, epochs=100, batch_size=32, shuffle=True, verbose=0)\n",
    "# Extraction de l'encodeur seul pour récupérer les signatures compressées\n",
    "encoder_modele = models.Model(inputs=input_layer, outputs=autoencoder.get_layer(\"couche_latente\").output)\n",
    "X_latent = encoder_modele.predict(X)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "bb52dedb-0949-4f8c-91c2-0a158dbf0ad3",
   "metadata": {},
   "source": [
    "Une fois l'apprentissage terminé, nous disposons d'un espace latent que l'on peut explorer et exploiter pour construire une classification. Dans ce cadre non supervisé, le réseau de neurones ne produit pas directement les classes. Son rôle consiste plutôt à apprendre une représentation compacte des données en réduisant la dimensionnalité du problème. Cette démarche est conceptuellement proche de celle de l'ACP ou de l'ACM : dans les deux cas, il s'agit de résumer l'information contenue dans un grand nombre de variables à l'aide d'un nombre réduit de dimensions. La différence essentielle est qu'un autoencodeur est capable d'apprendre des relations non linéaires entre les variables. Pour autant, les questions méthodologiques classiques ne disparaissent pas. Il faut toujours déterminer le nombre de dimensions de l'espace latent, choisir une méthode de classification et définir le nombre de classes à retenir. Dans cet exemple, nous utilisons simplement l'algorithme K-means afin de partitionner l'espace latent en cinq groupes."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "id": "dc0c1948-d1b6-4795-a760-5aa528b2d784",
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "C:\\Users\\Serge\\anaconda3\\Lib\\site-packages\\sklearn\\cluster\\_kmeans.py:870: FutureWarning: The default value of `n_init` will change from 10 to 'auto' in 1.4. Set the value of `n_init` explicitly to suppress the warning\n",
      "  warnings.warn(\n",
      "C:\\Users\\Serge\\anaconda3\\Lib\\site-packages\\sklearn\\cluster\\_kmeans.py:1382: UserWarning: KMeans is known to have a memory leak on Windows with MKL, when there are less chunks than available threads. You can avoid it by setting the environment variable OMP_NUM_THREADS=6.\n",
      "  warnings.warn(\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "<Axes: >"
      ]
     },
     "execution_count": 6,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "from sklearn.cluster import KMeans\n",
    "\n",
    "kmeans = KMeans(n_clusters=5, random_state=42)\n",
    "labels = kmeans.fit_predict(X_latent)\n",
    "\n",
    "resultats = pd.DataFrame({'INSEE_COM': matrice_brute.index, 'cluster_id': labels})\n",
    "zones_classifiees = zones.merge(resultats, on='INSEE_COM', how='left')\n",
    "zones_classifiees.plot('cluster_id')"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "f3ce35dd-ed10-4d68-b9c4-762130f2a5d8",
   "metadata": {},
   "source": [
    "L'évaluation d'une classification obtenue par apprentissage non supervisé n'est pas une tâche simple. Dans un premier temps, il est possible d'examiner les résultats afin d'en apprécier la cohérence géographique et l'interprétabilité. Dans notre cas, les classes obtenues semblent faire apparaître une logique centre-périphérie relativement cohérente. Il est toutefois possible que la prise en compte explicite du nombre total d'équipements ait renforcé cette structure au détriment d'autres dimensions plus fines, hypothèse qu'il serait possible de tester statistiquement. Cette question est d'autant plus complexe à résoudre ici que les données utilisées n'ont fait l'objet d'aucune pondération préalable. Les 228 catégories d'équipements sont traitées de manière équivalente, indépendamment de leur importance fonctionnelle ou de leur rôle dans l'organisation territoriale. Dès lors, le réseau tend naturellement à apprendre les structures les plus saillantes présentes dans les données. Si certaines dimensions jugées importantes par l'analyste ne sont que faiblement représentées, elles risquent d'être moins visibles dans l'espace latent. N'hésitez pas à expérimenter différents paramètres, architectures ou transformations des données afin d'observer l'impact de ces choix sur les classifications produites."
   ]
  },
  {
   "cell_type": "markdown",
   "id": "f48a218a-2b65-409a-8b33-eaf756d59af3",
   "metadata": {},
   "source": [
    "<b>Dit plus simplement, on peut s'attendre naïvement que le réseau retrouve notre logique et produise des classifications attendues, mais ne pas l'aider rend la mission complexe voire impossible pour lui. Lorsqu'il n'est pas supervisé, logiquement il raisonne à sa façon. Si on veut qu'il raisonne comme nous sans données étiquetées, sans données initiales qu’il peut prédire et qui vont guider l'apprentissage, il faudra lui donner des indications dans le tableau de données.</b>"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "b03c8fca-8cd3-4fac-b3a4-bb9a2ac47aa1",
   "metadata": {},
   "source": [
    "J'espère que cet exemple montre que les réseaux de neurones ne remettent pas fondamentalement en cause les principes des classifications non supervisées. Ils constituent avant tout une alternative permettant d'apprendre des représentations non linéaires des données, tout en conservant la plupart des questionnements méthodologiques classiques. Leur intérêt apparaît notamment lorsque les relations entre variables sont complexes ou lorsque le volume de données devient très important. Comme pour tout modèle, il convient toutefois de vérifier la qualité de l'apprentissage. L'examen de la fonction de perte (loss) permet notamment de s'assurer de la convergence du réseau et de détecter d'éventuelles anomalies. L'analyse de l'espace latent constitue également un outil précieux. Dans notre exemple, l'information semble principalement portée par deux des cinq dimensions latentes. Cela suggère qu'un espace latent plus compact, limité à deux dimensions, pourrait sans doute être testé lors d'expérimentations ultérieures. Ce faible nombre laisse à penser qu'une variable, ici le nombre total d'équipements, écrase les autres."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "id": "f982d84a-e93a-4911-be5b-d4cef7ad3206",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([[2143.5483,    0.    ,    0.    , 3118.9468,    0.    ],\n",
       "       [1328.3812,    0.    ,    0.    , 1933.0895,    0.    ],\n",
       "       [1339.4637,    0.    ,    0.    , 1949.1862,    0.    ],\n",
       "       [1361.1276,    0.    ,    0.    , 1980.6943,    0.    ],\n",
       "       [1773.9767,    0.    ,    0.    , 2581.3057,    0.    ]],\n",
       "      dtype=float32)"
      ]
     },
     "execution_count": 7,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "X_latent[0:5,:]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "id": "646d9111-045d-4966-a8a1-c16eee371e93",
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "No artists with labels found to put in legend.  Note that artists whose label start with an underscore are ignored when legend() is called with no argument.\n"
     ]
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 800x500 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "import matplotlib.pyplot as plt\n",
    "\n",
    "loss_valeurs = history.history['loss']\n",
    "epoches = range(1, len(loss_valeurs) + 1)\n",
    "\n",
    "plt.figure(figsize=(8, 5))\n",
    "plt.plot(epoches, loss_valeurs, 'b-')\n",
    "plt.title(\"Fonction de Loss pendant l'entraînement\")\n",
    "plt.xlabel('Époques')\n",
    "plt.ylabel('Erreur (Loss)')\n",
    "plt.grid(True)\n",
    "plt.legend()\n",
    "\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "32f5a139-ac52-4a98-82fe-dbc03794bc86",
   "metadata": {},
   "source": [
    "## Embending : la base des LLM pour créer une sémantique territoriale"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "eeca013d-0fd3-4351-9e6d-9deeb8039ddf",
   "metadata": {},
   "source": [
    "Nous allons maintenant aborder la notion d'embedding textuel, qui constitue l'un des concepts fondamentaux à l'origine des grands modèles de langage (LLM). Si l'on a compris le principe des autoencodeurs et de l'espace latent, il devient relativement facile de saisir l'intuition derrière les embeddings. Dans les deux cas, l'objectif consiste à représenter des objets complexes à l'aide d'un nombre réduit de dimensions numériques. Un embedding peut ainsi être vu comme une représentation vectorielle apprise automatiquement par un réseau de neurones, autrement dit comme une forme particulière d'espace latent adaptée à un problème donné. L'encodeur est le mécanisme (la fonction) qui transforme les données jusqu'à l'espace latent  (l'espace géométrique appris par le modèle). L'embending ce sont les vecteurs finaux (la position d'un objet dans cet espace)."
   ]
  },
  {
   "cell_type": "markdown",
   "id": "260518a9-f153-4515-a3a5-83fefabca416",
   "metadata": {},
   "source": [
    "<b> Si l'on applique cette idée aux points d'intérêt, la problématique devient alors : « Dis-moi quels lieux t'entourent et je te dirai quel type de lieu tu es. » Cette approche introduit une dimension nouvelle dans l'analyse spatiale quantitative : elle cherche à faire émerger la sémantique des lieux à partir des configurations spatiales observées. Bien entendu, la géographie et la statistique spatiale disposaient déjà de nombreux outils pour étudier les relations entre objets géographiques. Cependant, ces méthodes n'étaient généralement pas formulées en termes de sémantique spatiale et ne visaient pas explicitement à apprendre une représentation du « sens » des lieux à partir de leur contexte. Selon moi, ces approches nouvelles sont potentiellement disruptives<b>."
   ]
  },
  {
   "cell_type": "markdown",
   "id": "15e3f8fc-24fb-4559-b9f0-d9d054e24c3c",
   "metadata": {},
   "source": [
    "L'approche CBOW (Continuous Bag of Words) consiste à prédire le mot central à partir du contexte. L'approche Skip-Gram consiste à prédire le contexte à partir du mot central. Dans les deux cas se pose une question importante : faut-il tenir compte de l'ordre exact des mots ou simplement des mots présents dans le voisinage ? Les modèles de type Word2Vec adoptent une réponse relativement simple. Ils considèrent principalement les mots présents dans une fenêtre de contexte donnée, sans chercher à modéliser finement leur organisation syntaxique. Cette approche s'est révélée remarquablement efficace pour apprendre des représentations sémantiques des mots. Néanmoins, des architectures plus récentes peuvent conserver explicitement l'information de position grâce à des couches supplémentaires ou à des mécanismes dédiés, permettant ainsi au réseau d'apprendre que certains voisins ont une influence différente selon leur position dans la séquence. En tant que géographe dire que la distance n'a pas d'intérêt est une position forte et très discutable. On donc va s'inspirer des approches plus récentes pour tenir compte de la distance, plus exactement ici, de l'ordre de proximité."
   ]
  },
  {
   "cell_type": "markdown",
   "id": "fce2d6e2-603b-4883-b835-6b40c4e9b35b",
   "metadata": {},
   "source": [
    "On commence par charger les <a href='https://sergelhomme.fr/data/IA.zip'>données</a> : 39 types d'équipements répartis dans le Val-de-Marne. On cherche ensuite à déterminer le voisinage des équipements. Pour cela, on va retenir les 10 plus proches voisins par ordre de proximité à l'aide d'un algorithme efficace KDtree(). Chaque équipement possède ainsi une liste ordonnée de 10 voisins correspondant à 10 types d'équipements. A noter qu'un point d'intérêt peut très bien avoir plusieurs voisins de même nature, par exemple ci-dessous le supermarché a deux parfumeries à proximité direct. "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "id": "5fd85375-986c-435c-b14c-53d287536fe8",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "             NOM_EQ                                      voisins_types\n",
      "0  PRESSING-LAVERIE  [PAPETERIE_ET_PRESSE, SUPERMARCHE, MAGASIN_OPT...\n",
      "1       SUPERMARCHE  [SUPERMARCHE, PARFUMERIE, VETERINAIRE, PARFUME...\n",
      "2          EPICERIE  [DENTISTE, BANQUE, BOULANGERIE, BANQUE, RECHAR...\n",
      "3          EPICERIE  [CRECHE, EPICERIE, MAGASIN_ELECTROMENAGER, MED...\n",
      "4          EPICERIE  [PRESSING-LAVERIE, MAGASIN_EQUIPEMENTS_DU_FOYE...\n"
     ]
    }
   ],
   "source": [
    "import geopandas as gpd\n",
    "import numpy as np\n",
    "from scipy.spatial import KDTree\n",
    "\n",
    "gdf = gpd.read_file(\"BPE_IDF_39_VDM.shp\")  \n",
    "coordonnees = np.array(list(zip(gdf.geometry.x, gdf.geometry.y))) #Extraction des coordonnées X, Y \n",
    "arbre = KDTree(coordonnees)\n",
    "distances, indices = arbre.query(coordonnees, k=11) #on cherche les 11 voisins les plus proches\n",
    "indices_voisins = indices[:, 1:]  #Extraction des types de POI à partir des indices trouvés\n",
    "types_codes = gdf['NOM_EQ'].values\n",
    "gdf['voisins_types'] = [list(types_codes[idx]) for idx in indices_voisins] # Création de la liste des types des 10 voisins pour chaque ligne\n",
    "print(gdf[['NOM_EQ', 'voisins_types']].head())"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "99ad8256-1a40-416a-b4e5-c424cf796ca4",
   "metadata": {},
   "source": [
    "Pour créer notre réseau, nous commençons par transformer les mots en identifiants entiers. Cette étape ne produit pas encore de représentation sémantique : elle permet simplement d'obtenir un format exploitable par la couche d'embedding. Celle-ci apprendra ensuite à associer à chaque mot un vecteur numérique reflétant son usage dans les données. Pour réaliser cet encodage initial, nous utilisons le LabelEncoder de sklearn."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "id": "747b3e61-76f4-4f30-8043-454a44b56c5f",
   "metadata": {},
   "outputs": [],
   "source": [
    "import numpy as np\n",
    "import tensorflow as tf\n",
    "from tensorflow.keras import layers, models\n",
    "from sklearn.preprocessing import LabelEncoder\n",
    "# On initialise l'encodeur de texte\n",
    "le = LabelEncoder()\n",
    "# On ajuste l'encodeur sur l'ensemble de tes 39 types de POI possibles\n",
    "le.fit(gdf['NOM_EQ'])\n",
    "\n",
    "taille_vocabulaire = len(le.classes_) \n",
    "# Encodage de la cible (la sortie : ce que le réseau doit deviner)\n",
    "y = le.transform(gdf['NOM_EQ'])\n",
    "# Encodage des entrées (les listes de 10 voisins)\n",
    "# On transforme chaque liste de chaînes de caractères en liste d'entiers\n",
    "X = np.array([le.transform(voisins) for voisins in gdf['voisins_types']])"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "a7e7e4ec-83b2-4dcc-89c0-3f32a99b10ac",
   "metadata": {},
   "source": [
    "Notre réseau débute par une couche d'embedding chargée d'apprendre une représentation vectorielle des mots. La dimension des embeddings est fixée à 16 ; ce choix reste empirique et sera évalué à partir des résultats obtenus. Les vecteurs produits sont ensuite aplatis (Flatten) afin que chaque position du contexte soit représentée par un ensemble distinct de variables dans les couches suivantes. Cette opération permet au réseau de prendre en compte la position des mots dans la séquence. Le modèle comporte ensuite deux couches cachées de respectivement 64 et 32 neurones utilisant une fonction d'activation de type LeakyReLU. Enfin, la couche de sortie possède autant de neurones qu'il existe de types d'activités à prédire et utilise une activation softmax. Cette dernière produit une distribution de probabilités sur l'ensemble des classes possibles ; lors de la prédiction, la classe retenue correspond à celle présentant la probabilité la plus élevée."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "id": "2a1f14dd-6988-4aa5-ab3a-c54514a35546",
   "metadata": {},
   "outputs": [],
   "source": [
    "# =====================================================================\n",
    "# ARCHITECTURE DU RÉSEAU DE NEURONES (AVEC EMBEDDING)\n",
    "# =====================================================================\n",
    "\n",
    "dim_embedding = 16  # Chaque POI sera représenté par un vecteur de 16 chiffres\n",
    "longueur_sequence = 10  # On regarde les 10 voisins les plus proches\n",
    "\n",
    "model = models.Sequential([\n",
    "    # Couche d'Entrée : Reçoit les listes de 10 indices de voisins\n",
    "    layers.Input(shape=(longueur_sequence,)),\n",
    "    # Couche d'Embedding : Transforme (1300, 10) en (1300, 10, 16) elle crée l'espace sémantique de tes types d'équipements\n",
    "    layers.Embedding(input_dim=taille_vocabulaire, output_dim=dim_embedding),\n",
    "    # Couche de Flatten : On \"aplatit\" la matrice (10 * 16 = 160 neurones) pour la passer aux couches denses suivantes, en préservant l'ordre\n",
    "    layers.Flatten(),\n",
    "    # Couches cachées non linéaires pour capter les synergies de voisinage\n",
    "    layers.Dense(64),\n",
    "    layers.LeakyReLU(negative_slope=0.01),\n",
    "    layers.Dropout(0.2), # Évite le surapprentissage\n",
    "    layers.Dense(32),\n",
    "    layers.LeakyReLU(negative_slope=0.01),\n",
    "    # Couche de Sortie : 39 neurones avec Softmax pour obtenir les probabilités\n",
    "    layers.Dense(taille_vocabulaire, activation='softmax')\n",
    "])"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "1c3b6273-0f6f-42e5-93e1-8f73b7eb7c3b",
   "metadata": {},
   "source": [
    "On compile le modèle, toujours avec l'optimiseur adam."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "id": "e58e0a48-4707-46ad-ba80-11e94681713f",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"font-weight: bold\">Model: \"sequential\"</span>\n",
       "</pre>\n"
      ],
      "text/plain": [
       "\u001b[1mModel: \"sequential\"\u001b[0m\n"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [
       "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\">┏━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━┓\n",
       "┃<span style=\"font-weight: bold\"> Layer (type)                         </span>┃<span style=\"font-weight: bold\"> Output Shape                </span>┃<span style=\"font-weight: bold\">         Param # </span>┃\n",
       "┡━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━┩\n",
       "│ embedding (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">Embedding</span>)                │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">10</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">16</span>)              │             <span style=\"color: #00af00; text-decoration-color: #00af00\">624</span> │\n",
       "├──────────────────────────────────────┼─────────────────────────────┼─────────────────┤\n",
       "│ flatten (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">Flatten</span>)                    │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">160</span>)                 │               <span style=\"color: #00af00; text-decoration-color: #00af00\">0</span> │\n",
       "├──────────────────────────────────────┼─────────────────────────────┼─────────────────┤\n",
       "│ dense_3 (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">Dense</span>)                      │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">64</span>)                  │          <span style=\"color: #00af00; text-decoration-color: #00af00\">10,304</span> │\n",
       "├──────────────────────────────────────┼─────────────────────────────┼─────────────────┤\n",
       "│ leaky_re_lu (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">LeakyReLU</span>)              │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">64</span>)                  │               <span style=\"color: #00af00; text-decoration-color: #00af00\">0</span> │\n",
       "├──────────────────────────────────────┼─────────────────────────────┼─────────────────┤\n",
       "│ dropout (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">Dropout</span>)                    │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">64</span>)                  │               <span style=\"color: #00af00; text-decoration-color: #00af00\">0</span> │\n",
       "├──────────────────────────────────────┼─────────────────────────────┼─────────────────┤\n",
       "│ dense_4 (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">Dense</span>)                      │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">32</span>)                  │           <span style=\"color: #00af00; text-decoration-color: #00af00\">2,080</span> │\n",
       "├──────────────────────────────────────┼─────────────────────────────┼─────────────────┤\n",
       "│ leaky_re_lu_1 (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">LeakyReLU</span>)            │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">32</span>)                  │               <span style=\"color: #00af00; text-decoration-color: #00af00\">0</span> │\n",
       "├──────────────────────────────────────┼─────────────────────────────┼─────────────────┤\n",
       "│ dense_5 (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">Dense</span>)                      │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">39</span>)                  │           <span style=\"color: #00af00; text-decoration-color: #00af00\">1,287</span> │\n",
       "└──────────────────────────────────────┴─────────────────────────────┴─────────────────┘\n",
       "</pre>\n"
      ],
      "text/plain": [
       "┏━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━┓\n",
       "┃\u001b[1m \u001b[0m\u001b[1mLayer (type)                        \u001b[0m\u001b[1m \u001b[0m┃\u001b[1m \u001b[0m\u001b[1mOutput Shape               \u001b[0m\u001b[1m \u001b[0m┃\u001b[1m \u001b[0m\u001b[1m        Param #\u001b[0m\u001b[1m \u001b[0m┃\n",
       "┡━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━┩\n",
       "│ embedding (\u001b[38;5;33mEmbedding\u001b[0m)                │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m10\u001b[0m, \u001b[38;5;34m16\u001b[0m)              │             \u001b[38;5;34m624\u001b[0m │\n",
       "├──────────────────────────────────────┼─────────────────────────────┼─────────────────┤\n",
       "│ flatten (\u001b[38;5;33mFlatten\u001b[0m)                    │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m160\u001b[0m)                 │               \u001b[38;5;34m0\u001b[0m │\n",
       "├──────────────────────────────────────┼─────────────────────────────┼─────────────────┤\n",
       "│ dense_3 (\u001b[38;5;33mDense\u001b[0m)                      │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m64\u001b[0m)                  │          \u001b[38;5;34m10,304\u001b[0m │\n",
       "├──────────────────────────────────────┼─────────────────────────────┼─────────────────┤\n",
       "│ leaky_re_lu (\u001b[38;5;33mLeakyReLU\u001b[0m)              │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m64\u001b[0m)                  │               \u001b[38;5;34m0\u001b[0m │\n",
       "├──────────────────────────────────────┼─────────────────────────────┼─────────────────┤\n",
       "│ dropout (\u001b[38;5;33mDropout\u001b[0m)                    │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m64\u001b[0m)                  │               \u001b[38;5;34m0\u001b[0m │\n",
       "├──────────────────────────────────────┼─────────────────────────────┼─────────────────┤\n",
       "│ dense_4 (\u001b[38;5;33mDense\u001b[0m)                      │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m32\u001b[0m)                  │           \u001b[38;5;34m2,080\u001b[0m │\n",
       "├──────────────────────────────────────┼─────────────────────────────┼─────────────────┤\n",
       "│ leaky_re_lu_1 (\u001b[38;5;33mLeakyReLU\u001b[0m)            │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m32\u001b[0m)                  │               \u001b[38;5;34m0\u001b[0m │\n",
       "├──────────────────────────────────────┼─────────────────────────────┼─────────────────┤\n",
       "│ dense_5 (\u001b[38;5;33mDense\u001b[0m)                      │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m39\u001b[0m)                  │           \u001b[38;5;34m1,287\u001b[0m │\n",
       "└──────────────────────────────────────┴─────────────────────────────┴─────────────────┘\n"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [
       "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"font-weight: bold\"> Total params: </span><span style=\"color: #00af00; text-decoration-color: #00af00\">14,295</span> (55.84 KB)\n",
       "</pre>\n"
      ],
      "text/plain": [
       "\u001b[1m Total params: \u001b[0m\u001b[38;5;34m14,295\u001b[0m (55.84 KB)\n"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [
       "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"font-weight: bold\"> Trainable params: </span><span style=\"color: #00af00; text-decoration-color: #00af00\">14,295</span> (55.84 KB)\n",
       "</pre>\n"
      ],
      "text/plain": [
       "\u001b[1m Trainable params: \u001b[0m\u001b[38;5;34m14,295\u001b[0m (55.84 KB)\n"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [
       "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"font-weight: bold\"> Non-trainable params: </span><span style=\"color: #00af00; text-decoration-color: #00af00\">0</span> (0.00 B)\n",
       "</pre>\n"
      ],
      "text/plain": [
       "\u001b[1m Non-trainable params: \u001b[0m\u001b[38;5;34m0\u001b[0m (0.00 B)\n"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 1/80\n",
      "\u001b[1m273/273\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 3ms/step - accuracy: 0.0965 - loss: 3.4714 - val_accuracy: 0.2203 - val_loss: 2.9502\n",
      "Epoch 2/80\n",
      "\u001b[1m273/273\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 2ms/step - accuracy: 0.2255 - loss: 2.9622 - val_accuracy: 0.2464 - val_loss: 2.8439\n",
      "Epoch 3/80\n",
      "\u001b[1m273/273\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 2ms/step - accuracy: 0.2432 - loss: 2.8638 - val_accuracy: 0.2639 - val_loss: 2.8053\n",
      "Epoch 4/80\n",
      "\u001b[1m273/273\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 2ms/step - accuracy: 0.2732 - loss: 2.7909 - val_accuracy: 0.2680 - val_loss: 2.7879\n",
      "Epoch 5/80\n",
      "\u001b[1m273/273\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 2ms/step - accuracy: 0.2796 - loss: 2.7396 - val_accuracy: 0.2726 - val_loss: 2.7870\n",
      "Epoch 6/80\n",
      "\u001b[1m273/273\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 2ms/step - accuracy: 0.2899 - loss: 2.6990 - val_accuracy: 0.2827 - val_loss: 2.7730\n",
      "Epoch 7/80\n",
      "\u001b[1m273/273\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 2ms/step - accuracy: 0.2981 - loss: 2.6906 - val_accuracy: 0.2845 - val_loss: 2.7792\n",
      "Epoch 8/80\n",
      "\u001b[1m273/273\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 2ms/step - accuracy: 0.3061 - loss: 2.6516 - val_accuracy: 0.2799 - val_loss: 2.7844\n",
      "Epoch 9/80\n",
      "\u001b[1m273/273\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 2ms/step - accuracy: 0.2945 - loss: 2.6471 - val_accuracy: 0.2887 - val_loss: 2.7822\n",
      "Epoch 10/80\n",
      "\u001b[1m273/273\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 2ms/step - accuracy: 0.3059 - loss: 2.6172 - val_accuracy: 0.2896 - val_loss: 2.7854\n",
      "Epoch 11/80\n",
      "\u001b[1m273/273\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 2ms/step - accuracy: 0.3155 - loss: 2.5588 - val_accuracy: 0.2900 - val_loss: 2.7905\n",
      "Epoch 12/80\n",
      "\u001b[1m273/273\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 3ms/step - accuracy: 0.3138 - loss: 2.5600 - val_accuracy: 0.2887 - val_loss: 2.8060\n",
      "Epoch 13/80\n",
      "\u001b[1m273/273\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 3ms/step - accuracy: 0.3178 - loss: 2.5538 - val_accuracy: 0.2919 - val_loss: 2.8158\n",
      "Epoch 14/80\n",
      "\u001b[1m273/273\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 2ms/step - accuracy: 0.3244 - loss: 2.5331 - val_accuracy: 0.2910 - val_loss: 2.8161\n",
      "Epoch 15/80\n",
      "\u001b[1m273/273\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 2ms/step - accuracy: 0.3251 - loss: 2.4967 - val_accuracy: 0.2900 - val_loss: 2.8198\n",
      "Epoch 16/80\n",
      "\u001b[1m273/273\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 2ms/step - accuracy: 0.3244 - loss: 2.4801 - val_accuracy: 0.2896 - val_loss: 2.8287\n",
      "Epoch 17/80\n",
      "\u001b[1m273/273\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 2ms/step - accuracy: 0.3384 - loss: 2.4759 - val_accuracy: 0.2873 - val_loss: 2.8482\n",
      "Epoch 18/80\n",
      "\u001b[1m273/273\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 2ms/step - accuracy: 0.3333 - loss: 2.4607 - val_accuracy: 0.2965 - val_loss: 2.8514\n",
      "Epoch 19/80\n",
      "\u001b[1m273/273\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 2ms/step - accuracy: 0.3397 - loss: 2.4333 - val_accuracy: 0.2919 - val_loss: 2.8545\n",
      "Epoch 20/80\n",
      "\u001b[1m273/273\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 2ms/step - accuracy: 0.3328 - loss: 2.4510 - val_accuracy: 0.2896 - val_loss: 2.8711\n",
      "Epoch 21/80\n",
      "\u001b[1m273/273\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 2ms/step - accuracy: 0.3482 - loss: 2.3911 - val_accuracy: 0.2887 - val_loss: 2.8786\n",
      "Epoch 22/80\n",
      "\u001b[1m273/273\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 2ms/step - accuracy: 0.3479 - loss: 2.3864 - val_accuracy: 0.2887 - val_loss: 2.8987\n",
      "Epoch 23/80\n",
      "\u001b[1m273/273\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 2ms/step - accuracy: 0.3517 - loss: 2.3284 - val_accuracy: 0.2887 - val_loss: 2.8922\n",
      "Epoch 24/80\n",
      "\u001b[1m273/273\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 2ms/step - accuracy: 0.3468 - loss: 2.3488 - val_accuracy: 0.2855 - val_loss: 2.9118\n",
      "Epoch 25/80\n",
      "\u001b[1m273/273\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 2ms/step - accuracy: 0.3512 - loss: 2.3668 - val_accuracy: 0.2786 - val_loss: 2.9232\n",
      "Epoch 26/80\n",
      "\u001b[1m273/273\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 2ms/step - accuracy: 0.3505 - loss: 2.3386 - val_accuracy: 0.2868 - val_loss: 2.9373\n",
      "Epoch 27/80\n",
      "\u001b[1m273/273\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 2ms/step - accuracy: 0.3636 - loss: 2.2864 - val_accuracy: 0.2836 - val_loss: 2.9430\n",
      "Epoch 28/80\n",
      "\u001b[1m273/273\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 3ms/step - accuracy: 0.3488 - loss: 2.2975 - val_accuracy: 0.2822 - val_loss: 2.9606\n",
      "Epoch 29/80\n",
      "\u001b[1m273/273\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 2ms/step - accuracy: 0.3584 - loss: 2.2617 - val_accuracy: 0.2804 - val_loss: 2.9782\n",
      "Epoch 30/80\n",
      "\u001b[1m273/273\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 3ms/step - accuracy: 0.3588 - loss: 2.2666 - val_accuracy: 0.2804 - val_loss: 2.9726\n",
      "Epoch 31/80\n",
      "\u001b[1m273/273\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 2ms/step - accuracy: 0.3651 - loss: 2.2498 - val_accuracy: 0.2735 - val_loss: 2.9831\n",
      "Epoch 32/80\n",
      "\u001b[1m273/273\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 3ms/step - accuracy: 0.3667 - loss: 2.2410 - val_accuracy: 0.2763 - val_loss: 3.0001\n",
      "Epoch 33/80\n",
      "\u001b[1m273/273\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 3ms/step - accuracy: 0.3662 - loss: 2.2340 - val_accuracy: 0.2735 - val_loss: 3.0175\n",
      "Epoch 34/80\n",
      "\u001b[1m273/273\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 2ms/step - accuracy: 0.3698 - loss: 2.2167 - val_accuracy: 0.2726 - val_loss: 3.0143\n",
      "Epoch 35/80\n",
      "\u001b[1m273/273\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 2ms/step - accuracy: 0.3657 - loss: 2.2137 - val_accuracy: 0.2749 - val_loss: 3.0218\n",
      "Epoch 36/80\n",
      "\u001b[1m273/273\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 3ms/step - accuracy: 0.3680 - loss: 2.1877 - val_accuracy: 0.2772 - val_loss: 3.0317\n",
      "Epoch 37/80\n",
      "\u001b[1m273/273\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 2ms/step - accuracy: 0.3693 - loss: 2.1901 - val_accuracy: 0.2731 - val_loss: 3.0503\n",
      "Epoch 38/80\n",
      "\u001b[1m273/273\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 2ms/step - accuracy: 0.3714 - loss: 2.1859 - val_accuracy: 0.2671 - val_loss: 3.0495\n",
      "Epoch 39/80\n",
      "\u001b[1m273/273\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 2ms/step - accuracy: 0.3685 - loss: 2.1877 - val_accuracy: 0.2740 - val_loss: 3.0703\n",
      "Epoch 40/80\n",
      "\u001b[1m273/273\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 3ms/step - accuracy: 0.3692 - loss: 2.1596 - val_accuracy: 0.2676 - val_loss: 3.0818\n",
      "Epoch 41/80\n",
      "\u001b[1m273/273\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 2ms/step - accuracy: 0.3822 - loss: 2.1425 - val_accuracy: 0.2717 - val_loss: 3.1019\n",
      "Epoch 42/80\n",
      "\u001b[1m273/273\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 2ms/step - accuracy: 0.3776 - loss: 2.1240 - val_accuracy: 0.2694 - val_loss: 3.0949\n",
      "Epoch 43/80\n",
      "\u001b[1m273/273\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 2ms/step - accuracy: 0.3863 - loss: 2.1088 - val_accuracy: 0.2754 - val_loss: 3.0972\n",
      "Epoch 44/80\n",
      "\u001b[1m273/273\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 2ms/step - accuracy: 0.3935 - loss: 2.1023 - val_accuracy: 0.2698 - val_loss: 3.1182\n",
      "Epoch 45/80\n",
      "\u001b[1m273/273\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 2ms/step - accuracy: 0.3907 - loss: 2.0767 - val_accuracy: 0.2680 - val_loss: 3.1231\n",
      "Epoch 46/80\n",
      "\u001b[1m273/273\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 2ms/step - accuracy: 0.3791 - loss: 2.1196 - val_accuracy: 0.2694 - val_loss: 3.1355\n",
      "Epoch 47/80\n",
      "\u001b[1m273/273\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 3ms/step - accuracy: 0.3849 - loss: 2.1118 - val_accuracy: 0.2657 - val_loss: 3.1419\n",
      "Epoch 48/80\n",
      "\u001b[1m273/273\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 3ms/step - accuracy: 0.3897 - loss: 2.0780 - val_accuracy: 0.2735 - val_loss: 3.1559\n",
      "Epoch 49/80\n",
      "\u001b[1m273/273\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 4ms/step - accuracy: 0.3987 - loss: 2.0493 - val_accuracy: 0.2767 - val_loss: 3.1606\n",
      "Epoch 50/80\n",
      "\u001b[1m273/273\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 3ms/step - accuracy: 0.3938 - loss: 2.0710 - val_accuracy: 0.2698 - val_loss: 3.1684\n",
      "Epoch 51/80\n",
      "\u001b[1m273/273\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 3ms/step - accuracy: 0.3965 - loss: 2.0745 - val_accuracy: 0.2653 - val_loss: 3.1989\n",
      "Epoch 52/80\n",
      "\u001b[1m273/273\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 3ms/step - accuracy: 0.3974 - loss: 2.0601 - val_accuracy: 0.2657 - val_loss: 3.2007\n",
      "Epoch 53/80\n",
      "\u001b[1m273/273\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 3ms/step - accuracy: 0.3927 - loss: 2.0504 - val_accuracy: 0.2653 - val_loss: 3.2135\n",
      "Epoch 54/80\n",
      "\u001b[1m273/273\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 3ms/step - accuracy: 0.3985 - loss: 2.0331 - val_accuracy: 0.2648 - val_loss: 3.2229\n",
      "Epoch 55/80\n",
      "\u001b[1m273/273\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 2ms/step - accuracy: 0.4036 - loss: 2.0310 - val_accuracy: 0.2588 - val_loss: 3.2210\n",
      "Epoch 56/80\n",
      "\u001b[1m273/273\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 3ms/step - accuracy: 0.4016 - loss: 2.0156 - val_accuracy: 0.2643 - val_loss: 3.2266\n",
      "Epoch 57/80\n",
      "\u001b[1m273/273\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 3ms/step - accuracy: 0.3972 - loss: 2.0318 - val_accuracy: 0.2584 - val_loss: 3.2426\n",
      "Epoch 58/80\n",
      "\u001b[1m273/273\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 2ms/step - accuracy: 0.4057 - loss: 2.0146 - val_accuracy: 0.2579 - val_loss: 3.2484\n",
      "Epoch 59/80\n",
      "\u001b[1m273/273\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 2ms/step - accuracy: 0.4093 - loss: 2.0158 - val_accuracy: 0.2570 - val_loss: 3.2626\n",
      "Epoch 60/80\n",
      "\u001b[1m273/273\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 2ms/step - accuracy: 0.4141 - loss: 1.9742 - val_accuracy: 0.2593 - val_loss: 3.2574\n",
      "Epoch 61/80\n",
      "\u001b[1m273/273\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 3ms/step - accuracy: 0.4034 - loss: 2.0067 - val_accuracy: 0.2565 - val_loss: 3.2872\n",
      "Epoch 62/80\n",
      "\u001b[1m273/273\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 2ms/step - accuracy: 0.3948 - loss: 2.0064 - val_accuracy: 0.2634 - val_loss: 3.2895\n",
      "Epoch 63/80\n",
      "\u001b[1m273/273\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 3ms/step - accuracy: 0.4039 - loss: 1.9841 - val_accuracy: 0.2630 - val_loss: 3.3148\n",
      "Epoch 64/80\n",
      "\u001b[1m273/273\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 2ms/step - accuracy: 0.4171 - loss: 1.9755 - val_accuracy: 0.2593 - val_loss: 3.3081\n",
      "Epoch 65/80\n",
      "\u001b[1m273/273\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 3ms/step - accuracy: 0.4092 - loss: 1.9735 - val_accuracy: 0.2611 - val_loss: 3.3229\n",
      "Epoch 66/80\n",
      "\u001b[1m273/273\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 2ms/step - accuracy: 0.4140 - loss: 1.9557 - val_accuracy: 0.2575 - val_loss: 3.3240\n",
      "Epoch 67/80\n",
      "\u001b[1m273/273\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 3ms/step - accuracy: 0.4111 - loss: 1.9604 - val_accuracy: 0.2602 - val_loss: 3.3333\n",
      "Epoch 68/80\n",
      "\u001b[1m273/273\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 3ms/step - accuracy: 0.4179 - loss: 1.9623 - val_accuracy: 0.2666 - val_loss: 3.3289\n",
      "Epoch 69/80\n",
      "\u001b[1m273/273\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 2ms/step - accuracy: 0.4192 - loss: 1.9498 - val_accuracy: 0.2542 - val_loss: 3.3466\n",
      "Epoch 70/80\n",
      "\u001b[1m273/273\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 3ms/step - accuracy: 0.4078 - loss: 1.9655 - val_accuracy: 0.2616 - val_loss: 3.3689\n",
      "Epoch 71/80\n",
      "\u001b[1m273/273\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 2ms/step - accuracy: 0.4228 - loss: 1.9100 - val_accuracy: 0.2593 - val_loss: 3.3618\n",
      "Epoch 72/80\n",
      "\u001b[1m273/273\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 2ms/step - accuracy: 0.4105 - loss: 1.9538 - val_accuracy: 0.2616 - val_loss: 3.3699\n",
      "Epoch 73/80\n",
      "\u001b[1m273/273\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 3ms/step - accuracy: 0.4138 - loss: 1.9237 - val_accuracy: 0.2630 - val_loss: 3.3929\n",
      "Epoch 74/80\n",
      "\u001b[1m273/273\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 3ms/step - accuracy: 0.4033 - loss: 1.9628 - val_accuracy: 0.2653 - val_loss: 3.3901\n",
      "Epoch 75/80\n",
      "\u001b[1m273/273\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 3ms/step - accuracy: 0.4195 - loss: 1.9256 - val_accuracy: 0.2602 - val_loss: 3.3990\n",
      "Epoch 76/80\n",
      "\u001b[1m273/273\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 2ms/step - accuracy: 0.4150 - loss: 1.9377 - val_accuracy: 0.2598 - val_loss: 3.3970\n",
      "Epoch 77/80\n",
      "\u001b[1m273/273\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 3ms/step - accuracy: 0.4256 - loss: 1.9156 - val_accuracy: 0.2538 - val_loss: 3.4067\n",
      "Epoch 78/80\n",
      "\u001b[1m273/273\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 2ms/step - accuracy: 0.4193 - loss: 1.8860 - val_accuracy: 0.2570 - val_loss: 3.4124\n",
      "Epoch 79/80\n",
      "\u001b[1m273/273\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 3ms/step - accuracy: 0.4273 - loss: 1.9008 - val_accuracy: 0.2588 - val_loss: 3.4357\n",
      "Epoch 80/80\n",
      "\u001b[1m273/273\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 3ms/step - accuracy: 0.4223 - loss: 1.9238 - val_accuracy: 0.2542 - val_loss: 3.4546\n"
     ]
    }
   ],
   "source": [
    "# 'sparse_categorical_crossentropy' est parfaite car 'y' contient des entiers (0 à 38) et non des vecteurs binaires (One-Hot encoded)\n",
    "model.compile(\n",
    "    optimizer='adam',\n",
    "    loss='sparse_categorical_crossentropy',\n",
    "    metrics=['accuracy']\n",
    ")\n",
    "model.summary()\n",
    "\n",
    "history = model.fit(\n",
    "    X, y, \n",
    "    epochs=80, \n",
    "    batch_size=32, \n",
    "    validation_split=0.2, # Garde 20% des données pour tester la généralisation\n",
    "    shuffle=True\n",
    ")"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "1e23830c-ea34-4c01-855b-7eaecab26c51",
   "metadata": {},
   "source": [
    "Bien entendu, on peut observer l'évolution de la fonction de loss pour vérifier que l'apprentissage s'est bien passé. On peut aussi regarder ce que produit le modèle sur certains exemples :"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "id": "91762f3d-b471-47d4-bf8f-08e14c46c48c",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "COLLEGE\n",
      "['TERRAINS_DE_JEUX', 'PRESSING-LAVERIE', 'RECHARGE_VE', 'ECOLE_PRIMAIRE', 'ACCUEIL_DE_LOISIR', 'RECHARGE_VE', 'VETERINAIRE', 'ACCUEIL_DE_LOISIR', 'ACCUEIL_DE_LOISIR', 'RECHARGE_VE']\n",
      "\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 94ms/step\n",
      "[8.3928734e-02 3.0777149e-03 2.1206073e-03 2.1306642e-03 6.1625942e-05\n",
      " 2.8441139e-02 6.5922454e-02 3.0475393e-01 8.4460163e-03 1.5319550e-02\n",
      " 7.4578857e-04 8.4470175e-03 2.6921999e-02 1.3197457e-02 2.7266217e-03\n",
      " 3.8608789e-04 1.8879889e-04 2.2408354e-01 1.9120071e-05 2.1077096e-03\n",
      " 2.9752188e-04 8.0991845e-04 8.2608581e-02 1.2611956e-02 1.1769765e-04\n",
      " 5.9316467e-06 4.8209622e-05 2.1286301e-04 6.1705210e-03 5.7809381e-04\n",
      " 3.6112575e-05 8.0829486e-03 7.7409162e-03 4.5228195e-03 3.8374970e-03\n",
      " 4.9597099e-03 5.0288141e-03 6.8905674e-02 3.9757942e-04]\n",
      "Top 10 des prédictions pour ce contexte :\n",
      "  - COLLEGE              : 30.5%\n",
      "  - GYMNASES             : 22.4%\n",
      "  - ACCUEIL_DE_LOISIR    : 8.4%\n",
      "  - MAGASIN_DE_VETEMENTS : 8.3%\n",
      "  - TERRAINS_DE_JEUX     : 6.9%\n",
      "  - BOULANGERIE          : 6.6%\n",
      "  - BOUCHERIE            : 2.8%\n",
      "  - ECOLE_PRIMAIRE       : 2.7%\n",
      "  - CRECHE               : 1.5%\n",
      "  - ELEMENTAIRE          : 1.3%\n"
     ]
    }
   ],
   "source": [
    "nouvelle_liste_voisins = gdf['voisins_types'][8]\n",
    "print(gdf['NOM_EQ'][8])\n",
    "print(nouvelle_liste_voisins)\n",
    "probabilites = model.predict(np.array([le.transform(nouvelle_liste_voisins)]))[0]\n",
    "print(probabilites)\n",
    "\n",
    "indices_top10 = np.argsort(probabilites)[::-1][:10]\n",
    "print(\"Top 10 des prédictions pour ce contexte :\")\n",
    "for idx in indices_top10:\n",
    "    print(f\"  - {le.classes_[idx]:<20} : {probabilites[idx]*100:.1f}%\")"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "4d0c6d46-2d39-4ba2-89ae-2f6d9890a97c",
   "metadata": {},
   "source": [
    "Comme attendu, le réseau identifie correctement un collège à partir des types d'équipements présents dans son voisinage. Le modèle semble avoir appris qu'un environnement associant notamment des terrains de jeu, une école primaire ou encore un accueil de loisirs est fréquemment lié à la présence d'autres équipements d'enseignement. L'examen des autres probabilités prédites est également instructif pour évaluer la cohérence du modèle. En effet, lorsqu'un grand nombre de catégories est considéré, il est souvent plus pertinent d'analyser les prédictions les plus probables que de se limiter à la seule classe majoritaire. Dans notre exemple, le modèle attribue également des probabilités relativement élevées à un gymnase, à un terrain de jeu ou à un accueil de loisir. Ces équipements partagent des caractéristiques fonctionnelles proches et apparaissent fréquemment dans des contextes similaires. À l'inverse, des activités telles qu'une agence de voyage ou un magasin d'électroménager reçoivent des probabilités très faibles. La distribution des probabilités suggère ainsi que le modèle a appris des associations spatiales cohérentes entre les différents types d'équipements. Enfin, les embeddings peuvent être projetés dans un espace de faible dimension à l'aide d'une ACP afin d'explorer les proximités apprises par le réseau. Dans notre cas, l'un des principaux axes semble notamment distinguer des équipements majoritairement publics d'équipements davantage orientés vers les activités privées, ce qui constitue une structure particulièrement plausible au regard de l'organisation des territoires."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "id": "9c9d2352-5c20-4acd-ae48-d4395bd73367",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "L'Axe 1 explique 11.3% de la variance.\n",
      "L'Axe 2 explique 10.4% de la variance.\n"
     ]
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 1400x1000 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "L'Axe 3 explique 9.7% de la variance.\n",
      "L'Axe 4 explique 8.7% de la variance.\n"
     ]
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 1400x1000 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "import matplotlib.pyplot as plt\n",
    "from sklearn.decomposition import PCA\n",
    "import pandas as pd\n",
    "\n",
    "couche_embedding = model.layers[0]\n",
    "poids_embedding = couche_embedding.get_weights()[0]\n",
    "noms_poi = le.classes_\n",
    "df_embedding = pd.DataFrame(poids_embedding, index=noms_poi)\n",
    "\n",
    "pca = PCA(n_components=4) # ACP\n",
    "coords_pca = pca.fit_transform(df_embedding)\n",
    "variance_expliquee = pca.explained_variance_ratio_ * 100\n",
    "\n",
    "print(f\"L'Axe 1 explique {variance_expliquee[0]:.1f}% de la variance.\")\n",
    "print(f\"L'Axe 2 explique {variance_expliquee[1]:.1f}% de la variance.\")\n",
    "plt.figure(figsize=(14, 10))\n",
    "plt.scatter(coords_pca[:, 0], coords_pca[:, 1], color='blue', alpha=0.6, edgecolors='k', s=80)\n",
    "for i, nom in enumerate(noms_poi):\n",
    "    plt.text(coords_pca[i, 0] + 0.02, coords_pca[i, 1] + 0.02, nom, fontsize=10, alpha=0.8)\n",
    "plt.title(\"Espace sémantique des équipements\", fontsize=14, fontweight='bold')\n",
    "plt.xlabel(f\"Axe Principal 1 ({variance_expliquee[0]:.1f}%)\", fontsize=12)\n",
    "plt.ylabel(f\"Axe Principal 2 ({variance_expliquee[1]:.1f}%)\", fontsize=12)\n",
    "plt.grid(True, linestyle='--', alpha=0.5)\n",
    "# Centrage des axes sur 0\n",
    "plt.axhline(0, color='black', linewidth=0.8, linestyle='-')\n",
    "plt.axvline(0, color='black', linewidth=0.8, linestyle='-')\n",
    "plt.show()\n",
    "\n",
    "print(f\"L'Axe 3 explique {variance_expliquee[2]:.1f}% de la variance.\")\n",
    "print(f\"L'Axe 4 explique {variance_expliquee[3]:.1f}% de la variance.\")\n",
    "plt.figure(figsize=(14, 10))\n",
    "plt.scatter(coords_pca[:, 2], coords_pca[:, 3], color='blue', alpha=0.6, edgecolors='k', s=80)\n",
    "for i, nom in enumerate(noms_poi):\n",
    "    plt.text(coords_pca[i, 2] + 0.02, coords_pca[i, 3] + 0.02, nom, fontsize=10, alpha=0.8)\n",
    "plt.title(\"Espace sémantique des équipements\", fontsize=14, fontweight='bold')\n",
    "plt.xlabel(f\"Axe Principal 3 ({variance_expliquee[2]:.1f}%)\", fontsize=12)\n",
    "plt.ylabel(f\"Axe Principal 4 ({variance_expliquee[3]:.1f}%)\", fontsize=12)\n",
    "plt.grid(True, linestyle='--', alpha=0.5)\n",
    "# Centrage des axes sur 0\n",
    "plt.axhline(0, color='black', linewidth=0.8, linestyle='-')\n",
    "plt.axvline(0, color='black', linewidth=0.8, linestyle='-')\n",
    "plt.show()"
   ]
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3 (ipykernel)",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.11.7"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 5
}