🚗 Segmentation Multi-Classes
+Système de segmentation sémantique pour véhicules autonomes
+![](\"https://datascientest.com/wp-content/uploads/2021/05/u-net-architecture-1024x682.png\")
\n","\n","- **FPN (Feature Pyramid Network)** 🧭 \n"," FPN est une architecture conçue pour capturer efficacement des objets à **différentes échelles**, grâce à une combinaison intelligente de caractéristiques extraites à plusieurs résolutions.\n","\n"," Elle repose sur deux étapes principales :\n"," \n"," 1. 🔽 **Chemin descendant (bottom-up)** : \n"," L’image passe dans un encodeur profond (souvent un ResNet), qui génère des cartes de caractéristiques à plusieurs résolutions (`1/4`, `1/8`, `1/16`, `1/32`). \n"," Plus la résolution est petite, plus l’information est abstraite.\n"," \n"," 2. 🔼 **Chemin ascendant (top-down)** : \n"," Les cartes profondes sont **upsamplées progressivement** (x2 à chaque étape) et **fusionnées avec des cartes plus superficielles** à l’aide de *connexions latérales* (lateral connections). \n"," Cette fusion permet de combiner des informations **globales et locales** à chaque niveau de la pyramide.\n","\n"," Finalement, les différentes cartes sont combinées et passées dans un petit réseau convolutionnel (souvent un 3x3) pour produire la **prédiction finale**.\n","\n"," 👉 Cette approche rend FPN particulièrement efficace sur des images avec des objets de **tailles très variées**, comme en vision par ordinateur généraliste (urbain, aérien, etc.).\n","\n"," Le modèle a été historiquement entraîné avec ResNet50 et ResNet101 par l'équipe de chercheurs.\n","> **“We build FPN on top of ResNet-50/101 [21], which are used as backbone feature extractors.”** \n","> — *Feature Pyramid Networks for Object Detection*, \n","> Tsung-Yi Lin, Piotr Dollár, Ross Girshick, Kaiming He, Serge Belongie. \n","> *CVPR 2017* \n","> [arXiv:1612.03144](https://arxiv.org/abs/1612.03144)\n","\n","\n"," *(Le schéma ci-dessous illustre cette architecture dans le contexte original de la détection, mais le principe est le même en segmentation sémantique.)*\n","\n"," \n","\n","---\n","\n","### Choix des encodeurs 🧠\n","\n","Plusieurs encodeurs sont disponibles pour ces architectures : \n","- **Encodeur intégré** 🆕 : L’encodeur est entraîné en même temps que le décodeur, sans poids pré-entraînés. Pour U-Net, il a été développé *from scratch* mais pour FPN, il s'agira de ResNet34 (plus léger que ResNet50, le modèle de feature extraction originellement utilisé dans les travaux de recherche).\n","- **MobilenetV2 pré-entraîné** 🚀 : Un encodeur léger et efficace, pré-entraîné sur ImageNet, permettant un bon compromis entre rapidité et performance. \n","- **EfficientNetB0 pré-entraîné** ⚡ : Un encodeur plus puissant et performant, également basé sur un modèle pré-entraîné, utilisé pour tirer parti du transfert learning.\n","\n","ℹ️ Par défaut, la bibliothèque `segmentation_models` utilise **ResNet34 pré-entraîné sur ImageNet** comme encodeur si aucun autre n’est spécifié.\n","\n","---\n","\n","### Gel ou fine-tuning de l’encodeur ❄️🔥\n","\n","Lors de l’entraînement, deux stratégies sont possibles : \n","- **Encodage gelé** ❄️ : Les poids de l’encodeur pré-entraîné restent fixes, ce qui accélère l’entraînement et évite la modification de caractéristiques déjà apprises. \n","- **Encodage entraînable (fine-tuning)** 🔥 : Les poids de l’encodeur sont ajustés pour mieux s’adapter aux données spécifiques, ce qui peut améliorer les performances au prix d’un temps d’entraînement plus long.\n","\n","---\n","\n","### Stratégie recommandée pour l’expérimentation 🎯\n","\n","1. **Tester toutes les architectures et encodeurs avec l’encodeur gelé** ❄️ pour un benchmark rapide. \n","2. **Sélectionner la meilleure configuration** 🏆 selon les performances obtenues. \n","3. **Refaire l’entraînement avec l’encodeur dégelé (fine-tuning)** 🔥 pour optimiser les résultats finaux.\n","\n","Cette méthode permet de gagner du temps tout en explorant efficacement l’espace des modèles.\n","\n"]},{"cell_type":"markdown","metadata":{"id":"s5te0Fcwtx51"},"source":["### Création des modèles"]},{"cell_type":"code","execution_count":25,"metadata":{"colab":{"base_uri":"https://localhost:8080/"},"executionInfo":{"elapsed":12,"status":"ok","timestamp":1754980975025,"user":{"displayName":"Grégoire","userId":"01249298206117462766"},"user_tz":-120},"id":"8eHYsAmIRsIU","outputId":"60fe35f3-26b7-4a3d-e559-84804cd5e43f"},"outputs":[{"output_type":"stream","name":"stdout","text":["🔧 BACKBONES DISPONIBLES:\n","==================================================\n","⭐ RECOMMANDÉS pour voiture autonome (vitesse/précision):\n"," • mobilenetv2\n"," • efficientnetb0\n"," • resnet34\n"," • resnet50\n","\n","📋 TOUS DISPONIBLES (32):\n","\n"," resnet18 resnet34 resnet50 resnet101 \n"," resnet152 seresnet18 seresnet34 seresnet50 \n"," seresnet101 seresnet152 seresnext50 seresnext101 \n"," senet154 resnext50 resnext101 vgg16 \n"," vgg19 densenet121 densenet169 densenet201 \n"," inceptionresnetv2 inceptionv3 mobilenet mobilenetv2 \n"," efficientnetb0 efficientnetb1 efficientnetb2 efficientnetb3 \n"," efficientnetb4 efficientnetb5 efficientnetb6 efficientnetb7 \n"]}],"source":["def show_available_backbones():\n"," \"\"\"Affiche tous les backbones disponibles dans segmentation_models\"\"\"\n","\n"," # Essayer de récupérer la liste des backbones disponibles\n"," try:\n"," available_backbones = sm.get_available_backbones()\n","\n"," print(\"🔧 BACKBONES DISPONIBLES DANS SEGMENTATION_MODELS:\")\n"," print(\"=\" * 60)\n","\n"," # Grouper par famille\n"," families = {\n"," 'ResNet': [b for b in available_backbones if b.startswith('resnet')],\n"," 'SEResNet': [b for b in available_backbones if b.startswith('seresnet')],\n"," 'SEResNeXt': [b for b in available_backbones if b.startswith('seresnext')],\n"," 'ResNeXt': [b for b in available_backbones if b.startswith('resnext')],\n"," 'VGG': [b for b in available_backbones if b.startswith('vgg')],\n"," 'DenseNet': [b for b in available_backbones if b.startswith('densenet')],\n"," 'Inception': [b for b in available_backbones if 'inception' in b],\n"," 'MobileNet': [b for b in available_backbones if 'mobilenet' in b],\n"," 'EfficientNet': [b for b in available_backbones if 'efficientnet' in b],\n"," 'SENet': [b for b in available_backbones if b.startswith('senet')],\n"," }\n","\n"," for family, models in families.items():\n"," if models:\n"," print(f\"\\n📱 {family}:\")\n"," for model in sorted(models):\n"," print(f\" • {model}\")\n","\n"," except AttributeError:\n"," # Si la méthode n'existe pas, utiliser la liste connue\n"," known_backbones = [\n"," 'resnet18', 'resnet34', 'resnet50', 'resnet101', 'resnet152',\n"," 'seresnet18', 'seresnet34', 'seresnet50', 'seresnet101', 'seresnet152',\n"," 'seresnext50', 'seresnext101', 'senet154', 'resnext50', 'resnext101',\n"," 'vgg16', 'vgg19', 'densenet121', 'densenet169', 'densenet201',\n"," 'inceptionresnetv2', 'inceptionv3', 'mobilenet', 'mobilenetv2',\n"," 'efficientnetb0', 'efficientnetb1', 'efficientnetb2', 'efficientnetb3',\n"," 'efficientnetb4', 'efficientnetb5', 'efficientnetb6', 'efficientnetb7'\n"," ]\n","\n"," print(\"🔧 BACKBONES DISPONIBLES:\")\n"," print(\"=\" * 50)\n","\n"," # Recommandations pour voiture autonome\n"," recommended = ['mobilenetv2', 'efficientnetb0', 'resnet34', 'resnet50']\n","\n"," print(\"⭐ RECOMMANDÉS pour voiture autonome (vitesse/précision):\")\n"," for model in recommended:\n"," print(f\" • {model}\")\n","\n"," print(f\"\\n📋 TOUS DISPONIBLES ({len(known_backbones)}):\")\n"," for i, model in enumerate(known_backbones):\n"," if i % 4 == 0: # Nouvelle ligne tous les 4\n"," print()\n"," print(f\" {model:<20}\", end=\"\")\n"," print() # Ligne finale\n","\n","# Afficher les backbones disponibles\n","show_available_backbones()"]},{"cell_type":"code","execution_count":26,"metadata":{"id":"3xLKhC8KW5i7","executionInfo":{"status":"ok","timestamp":1754980975305,"user_tz":-120,"elapsed":136,"user":{"displayName":"Grégoire","userId":"01249298206117462766"}}},"outputs":[],"source":["# ================================================================\n","# GRILLE D'EXPÉRIMENTATIONS - SEGMENTATION SÉMANTIQUE CITYSCAPES\n","# ================================================================\n","\n","# Architectures et backbones à tester\n","model_variants = [\n"," # Baselines (sans backbone pré-entraîné)\n"," ('U-Net', None),\n"," ('FPN', None), # --> l'encodeur choisi sera alors vgg16\n"," # Modèles avec backbones pré-entraînés (encoders gelés)\n"," ('U-Net', 'mobilenetv2'),\n"," ('U-Net', 'efficientnetb0'),\n"," ('FPN', 'mobilenetv2'),\n"," ('FPN', 'efficientnetb0'),\n"," ('FPN', 'resnet34'),\n","]\n","\n","# Fine-tuning sélectif (encoders entraînables)\n","finetune_variants = [\n"," ('U-Net', 'mobilenetv2'),\n"," ('FPN', 'efficientnetb0'),\n"," ('FPN', 'resnet34'),\n","]\n","\n","# Paramètres d'entraînement\n","base_params = {\n"," 'learning_rate': 1e-4,\n"," 'epochs': 20,\n"," 'batch_size': 8,\n"," 'img_size': (224, 224)\n","}\n","\n","finetune_params = {\n"," 'learning_rate': 5e-5, # Plus petit pour le fine-tuning\n"," 'epochs': 30,\n"," 'batch_size': 8,\n"," 'img_size': (224, 224)\n","}\n","\n","# Génération des configurations\n","def generate_experiment_configs():\n"," \"\"\"Génère toutes les configurations d'expérimentations\"\"\"\n"," model_configs = []\n","\n"," # 1. Expériences de base (encodeurs gelés ou vanilla)\n"," for arch, encoder in model_variants:\n"," for augment in [False, True]:\n"," # ✅ Nom correct selon le type\n"," if encoder is None:\n"," config_name = f\"{arch}_vanilla_aug{augment}\"\n"," else:\n"," config_name = f\"{arch}_{encoder}_frozen_aug{augment}\"\n","\n"," model_configs.append({\n"," \"name\": config_name,\n"," \"architecture\": arch,\n"," \"encoder\": encoder,\n"," \"encoder_freeze\": encoder is not None,\n"," \"augmentation\": augment,\n"," \"params\": base_params.copy(),\n"," \"type\": \"baseline\" if encoder is None else \"frozen_pretrained\"\n"," })\n","\n"," # 2. Expériences de fine-tuning (encoders dégelés)\n"," for arch, encoder in finetune_variants:\n"," config_name = f\"{arch}_{encoder}_finetune_augTrue\"\n"," model_configs.append({\n"," \"name\": config_name,\n"," \"architecture\": arch,\n"," \"encoder\": encoder,\n"," \"encoder_freeze\": False, # False = dégelé pour fine-tuning\n"," \"augmentation\": True,\n"," \"params\": finetune_params.copy(),\n"," \"type\": \"finetuned_pretrained\"\n"," })\n","\n"," return model_configs"]},{"cell_type":"code","execution_count":27,"metadata":{"id":"X7wY6OqSZFha","executionInfo":{"status":"ok","timestamp":1754980975483,"user_tz":-120,"elapsed":7,"user":{"displayName":"Grégoire","userId":"01249298206117462766"}}},"outputs":[],"source":["def create_model_from_config(config, num_classes, verbose=False):\n"," \"\"\"\n"," Crée un modèle de segmentation à partir d'une configuration d'expérimentation.\n"," \"\"\"\n"," # === EXTRACTION DES PARAMÈTRES ===\n"," architecture = config['architecture']\n"," encoder = config['encoder']\n"," encoder_freeze = config['encoder_freeze']\n"," learning_rate = config['params']['learning_rate']\n"," img_size = config['params']['img_size']\n","\n"," # === MAPPING DES ARCHITECTURES ===\n"," model_classes = {\n"," 'U-Net': sm.Unet,\n"," 'FPN': sm.FPN,\n"," }\n","\n"," if architecture not in model_classes:\n"," raise ValueError(f\"Architecture '{architecture}' non supportée.\")\n","\n"," ModelClass = model_classes[architecture]\n"," input_shape = (*img_size, 3)\n","\n"," if verbose:\n"," print(f\"🔧 Création du modèle '{config['name']}'\")\n"," print(f\" Architecture: {architecture}\")\n"," print(f\" Encoder: {encoder or 'Aucun (vanilla)'}\")\n"," print(f\" Encoder gelé: {encoder_freeze}\")\n","\n"," try:\n"," if encoder is None:\n"," # === MODÈLE VANILLA ===\n"," model = ModelClass(\n"," classes=num_classes,\n"," activation='softmax',\n"," input_shape=input_shape,\n"," )\n"," else:\n"," # === MODÈLE AVEC BACKBONE ===\n"," model = ModelClass(\n"," backbone_name=encoder,\n"," encoder_weights='imagenet',\n"," encoder_freeze=encoder_freeze,\n"," classes=num_classes,\n"," activation='softmax',\n"," input_shape=input_shape,\n"," )\n","\n"," # === COMPILATION ===\n"," model.compile(\n"," optimizer=keras.optimizers.Adam(learning_rate=learning_rate),\n"," loss=sm.losses.categorical_focal_jaccard_loss,\n"," metrics=[sm.metrics.iou_score, 'accuracy']\n"," )\n","\n"," if verbose:\n"," total_params = model.count_params()\n"," trainable_params = sum([tf.keras.backend.count_params(w) for w in model.trainable_weights])\n"," frozen_params = total_params - trainable_params\n"," print(f\" 📊 Paramètres: {total_params:,} total | {trainable_params:,} entraînables | {frozen_params:,} gelés\")\n","\n"," return model\n","\n"," except Exception as e:\n"," print(f\"❌ Erreur: {e}\")\n"," raise"]},{"cell_type":"code","execution_count":28,"metadata":{"id":"4ROnd_8f9_W7","executionInfo":{"status":"ok","timestamp":1754980975757,"user_tz":-120,"elapsed":42,"user":{"displayName":"Grégoire","userId":"01249298206117462766"}}},"outputs":[],"source":["# Génération des configurations\n","experimental_configurations = generate_experiment_configs()"]},{"cell_type":"markdown","metadata":{"id":"GILXYzC_9jHX"},"source":["### Analyse et compréhension détaillée des modèles"]},{"cell_type":"markdown","metadata":{"id":"5yr47E0qbmmY"},"source":["L'objectif est simplement ici d'afficher des informations générales sur les modèles et notamment vérifier le gel des poids pour les modèles de feature extraction (backbone) qui fournissent des descripteurs aux modèles de segmentation et ce afin de pouvoir comparer les différences de performances."]},{"cell_type":"code","execution_count":null,"metadata":{"id":"qdsGkrVQe2mX"},"outputs":[],"source":["def analyze_model_comprehensive(config, num_classes, show_architecture=False, detailed_layers=False):\n"," \"\"\"\n"," Analyse exhaustive d'un modèle de segmentation avec affichage clair des couches gelées/dégelées\n","\n"," Args:\n"," config (dict): Configuration du modèle\n"," num_classes (int): Nombre de classes\n"," show_architecture (bool): Afficher model.summary()\n"," detailed_layers (bool): Afficher TOUTES les couches (sinon juste backbone + résumé)\n","\n"," Returns:\n"," dict: Résultats de l'analyse\n"," \"\"\"\n"," print(f\"\\n{'='*70}\")\n"," print(f\"🏗️ ANALYSE COMPLÈTE: {config['name']}\")\n"," print(f\"{'='*70}\")\n","\n"," # === 1. INFORMATIONS DE CONFIGURATION ===\n"," print(f\"📊 Configuration:\")\n"," print(f\" • Architecture: {config['architecture']}\")\n"," print(f\" • Backbone: {config['encoder'] or 'Aucun (vanilla)'}\")\n"," print(f\" • Encoder gelé: {'✅ Oui' if config.get('encoder_freeze', False) else '❌ Non'}\")\n"," print(f\" • Type: {config['type'].replace('_', ' ').title()}\")\n"," print(f\" • Augmentation: {'✅ Oui' if config['augmentation'] else '❌ Non'}\")\n"," print(f\" • Learning rate: {config['params']['learning_rate']}\")\n"," print(f\" • Époques prévues: {config['params']['epochs']}\")\n"," print(f\" • Taille images: {config['params']['img_size']}\")\n","\n"," try:\n"," # === 2. CRÉATION DU MODÈLE ===\n"," print(f\"\\n🔧 Création du modèle...\")\n"," model = create_model_from_config(config, num_classes, verbose=False)\n","\n"," # === 3. ANALYSE DES PARAMÈTRES ===\n"," total_params = model.count_params()\n"," trainable_params = sum([tf.keras.backend.count_params(w) for w in model.trainable_weights])\n"," frozen_params = total_params - trainable_params\n"," frozen_ratio = (frozen_params / total_params) * 100\n","\n"," print(f\"\\n🔢 Paramètres:\")\n"," print(f\" • Total: {total_params:,}\")\n"," print(f\" • Entraînables: {trainable_params:,} ({(trainable_params/total_params)*100:.1f}%)\")\n"," print(f\" • Gelés: {frozen_params:,} ({frozen_ratio:.1f}%)\")\n","\n"," # === 4. ANALYSE DES COUCHES ===\n"," backbone_layers = []\n"," other_layers = []\n"," backbone_total = 0\n"," backbone_frozen = 0\n","\n"," # Classifier les couches\n"," for layer in model.layers:\n"," layer_params = layer.count_params()\n"," is_backbone = False\n","\n"," if config['encoder']:\n"," backbone_keywords = [\n"," config['encoder'].lower(),\n"," 'encoder',\n"," 'backbone',\n"," 'feature'\n"," ]\n"," is_backbone = any(keyword in layer.name.lower() for keyword in backbone_keywords)\n","\n"," layer_info = {\n"," 'name': layer.name,\n"," 'params': layer_params,\n"," 'trainable': layer.trainable,\n"," 'is_backbone': is_backbone\n"," }\n","\n"," if is_backbone:\n"," backbone_layers.append(layer_info)\n"," backbone_total += layer_params\n"," if not layer.trainable:\n"," backbone_frozen += layer_params\n"," else:\n"," other_layers.append(layer_info)\n","\n"," # === 5. AFFICHAGE DES COUCHES ===\n"," if detailed_layers:\n"," # Affichage complet de toutes les couches\n"," print(f\"\\n🔍 TOUTES LES COUCHES (🟢 = Entraînable, 🔴 = Gelé):\")\n"," print(\"-\" * 70)\n","\n"," all_layers = backbone_layers + other_layers\n"," for i, layer_info in enumerate(all_layers):\n"," icon = \"🔴\" if not layer_info['trainable'] else \"🟢\"\n"," status = \"GELÉ\" if not layer_info['trainable'] else \"ENTRAÎNABLE\"\n","\n"," # Identifier le type de couche\n"," if layer_info['is_backbone']:\n"," layer_type = \" (BACKBONE)\"\n"," elif 'decoder' in layer_info['name'].lower():\n"," layer_type = \" (DECODER)\"\n"," elif 'classifier' in layer_info['name'].lower() or 'final' in layer_info['name'].lower():\n"," layer_type = \" (CLASSIFICATION)\"\n"," elif 'upsample' in layer_info['name'].lower():\n"," layer_type = \" (UPSAMPLING)\"\n"," else:\n"," layer_type = \"\"\n","\n"," print(f\"{i+1:2d}. {icon} {layer_info['name']:<40} {layer_info['params']:>8,} params - {status}{layer_type}\")\n","\n"," else:\n"," # Affichage résumé avec focus sur le backbone\n"," if backbone_layers:\n"," print(f\"\\n🔍 COUCHES DU BACKBONE (🟢 = Entraînable, 🔴 = Gelé):\")\n"," print(\"-\" * 60)\n","\n"," for layer_info in backbone_layers[:8]: # Limite à 8 pour lisibilité\n"," icon = \"🔴\" if not layer_info['trainable'] else \"🟢\"\n"," status = \"GELÉ\" if not layer_info['trainable'] else \"ENTRAÎNABLE\"\n"," print(f\" {icon} {layer_info['name']:<35} {layer_info['params']:>8,} params - {status}\")\n","\n"," if len(backbone_layers) > 8:\n"," print(f\" ... et {len(backbone_layers)-8} autres couches du backbone\")\n","\n"," # Résumé des autres couches\n"," other_frozen = sum(1 for layer in other_layers if not layer['trainable'])\n"," other_trainable = len(other_layers) - other_frozen\n"," print(f\"\\n🏗️ AUTRES COUCHES (résumé):\")\n"," print(f\" • {other_trainable} couches entraînables\")\n"," print(f\" • {other_frozen} couches gelées\")\n"," print(f\" • (Utilisez detailed_layers=True pour voir toutes les couches)\")\n","\n"," # === 6. RÉSUMÉ BACKBONE ===\n"," if backbone_layers:\n"," backbone_frozen_ratio = (backbone_frozen / backbone_total * 100) if backbone_total > 0 else 0\n"," print(f\"\\n📋 Résumé backbone:\")\n"," print(f\" • Total backbone: {backbone_total:,} params\")\n"," print(f\" • Backbone gelé: {backbone_frozen:,} params ({backbone_frozen_ratio:.1f}%)\")\n"," print(f\" • Backbone entraînable: {backbone_total - backbone_frozen:,} params ({100-backbone_frozen_ratio:.1f}%)\")\n","\n"," # === 7. VALIDATION DU GEL ===\n"," if config['encoder']:\n"," print(f\"\\n🎯 Validation du gel du backbone:\")\n","\n"," # Seuils adaptatifs selon le backbone\n"," backbone_name = config['encoder'].lower()\n"," if 'mobilenet' in backbone_name:\n"," expected_frozen_range = (20, 35)\n"," elif 'efficientnet' in backbone_name and 'b0' in backbone_name:\n"," expected_frozen_range = (15, 25)\n"," elif 'resnet' in backbone_name:\n"," expected_frozen_range = (18, 28)\n"," else:\n"," expected_frozen_range = (10, 30)\n","\n"," if config.get('encoder_freeze', False):\n"," # Modèle supposé gelé\n"," if frozen_ratio >= expected_frozen_range[0]:\n"," status = \"✅ OK\"\n"," message = f\"Backbone correctement gelé ({frozen_ratio:.1f}% ≥ {expected_frozen_range[0]}%)\"\n"," else:\n"," status = \"⚠️ SUSPECT\"\n"," message = f\"Gel potentiellement incomplet ({frozen_ratio:.1f}% < {expected_frozen_range[0]}%)\"\n"," else:\n"," # Modèle supposé dégelé\n"," if frozen_ratio < 5.0:\n"," status = \"✅ OK\"\n"," message = f\"Backbone correctement dégelé ({frozen_ratio:.1f}% < 5%)\"\n"," else:\n"," status = \"⚠️ SUSPECT\"\n"," message = f\"Gel résiduel inattendu ({frozen_ratio:.1f}% ≥ 5%)\"\n","\n"," print(f\" • Statut global: {status}\")\n"," print(f\" • {message}\")\n"," print(f\" • Plage attendue pour {config['encoder']}: {expected_frozen_range[0]}-{expected_frozen_range[1]}%\")\n","\n"," # Validation spécifique du backbone\n"," if backbone_total > 0:\n"," backbone_expected_frozen = config.get('encoder_freeze', False)\n"," backbone_actually_frozen = (backbone_frozen / backbone_total) > 0.9\n","\n"," if backbone_expected_frozen == backbone_actually_frozen:\n"," backbone_status = \"✅ COHÉRENT\"\n"," else:\n"," backbone_status = \"❌ INCOHÉRENT\"\n","\n"," print(f\" • Cohérence backbone: {backbone_status}\")\n"," print(f\" - Configuration: {'gelé' if backbone_expected_frozen else 'dégelé'}\")\n"," print(f\" - Réalité: {'gelé' if backbone_actually_frozen else 'dégelé'} ({backbone_frozen_ratio:.1f}%)\")\n","\n"," # === 8. ARCHITECTURE DÉTAILLÉE (optionnel) ===\n"," if show_architecture:\n"," print(f\"\\n🏛️ Architecture détaillée (model.summary()):\")\n"," print(\"-\" * 60)\n"," model.summary(line_length=80)\n","\n"," # === 9. RÉSULTATS POUR RETOUR ===\n"," results = {\n"," 'model': model,\n"," 'total_params': total_params,\n"," 'trainable_params': trainable_params,\n"," 'frozen_params': frozen_params,\n"," 'frozen_ratio': frozen_ratio,\n"," 'backbone_params': backbone_total,\n"," 'backbone_frozen': backbone_frozen,\n"," 'backbone_frozen_ratio': backbone_frozen_ratio if backbone_total > 0 else 0,\n"," 'status': 'success',\n"," 'validation': status if config['encoder'] else 'N/A'\n"," }\n","\n"," print(f\"\\n✅ Analyse terminée avec succès\")\n"," return results\n","\n"," except Exception as e:\n"," print(f\"\\n❌ Erreur lors de l'analyse: {e}\")\n"," return {\n"," 'model': None,\n"," 'status': 'error',\n"," 'error': str(e)\n"," }\n","\n"," finally:\n"," # Nettoyage mémoire\n"," if 'model' in locals():\n"," del model\n"," tf.keras.backend.clear_session()\n","\n","# === FONCTION SÉPARÉE POUR VOIR UNIQUEMENT LES COUCHES (différente maintenant) ===\n","def show_all_layers_status(config, num_classes):\n"," \"\"\"\n"," Fonction spécialisée pour afficher UNIQUEMENT les couches, de manière compacte\n"," \"\"\"\n"," print(f\"\\n🔍 COUCHES - {config['name']}\")\n"," print(\"=\" * 70)\n"," print(\"🟢 ENTRAÎNABLE | 🔴 GELÉ\")\n"," print(\"-\" * 70)\n","\n"," try:\n"," model = create_model_from_config(config, num_classes, verbose=False)\n","\n"," trainable_count = 0\n"," frozen_count = 0\n","\n"," for i, layer in enumerate(model.layers):\n"," layer_params = layer.count_params()\n"," icon = \"🔴\" if not layer.trainable else \"🟢\"\n","\n"," if layer.trainable:\n"," trainable_count += 1\n"," else:\n"," frozen_count += 1\n","\n"," # Type de couche (plus concis)\n"," layer_type = \"\"\n"," if config['encoder'] and any(keyword in layer.name.lower()\n"," for keyword in [config['encoder'].lower(), 'encoder', 'backbone']):\n"," layer_type = \"BACKBONE\"\n"," elif 'decoder' in layer.name.lower():\n"," layer_type = \"DECODER\"\n"," elif 'classifier' in layer.name.lower():\n"," layer_type = \"CLASSIF\"\n"," else:\n"," layer_type = \"AUTRE\"\n","\n"," print(f\"{i+1:2d}. {icon} {layer.name:<35} {layer_params:>8,} {layer_type}\")\n","\n"," print(\"-\" * 70)\n"," print(f\"📊 RÉSUMÉ: {trainable_count} entraînables, {frozen_count} gelés\")\n","\n"," except Exception as e:\n"," print(f\"❌ Erreur: {e}\")\n"," finally:\n"," if 'model' in locals():\n"," del model\n"," tf.keras.backend.clear_session()"]},{"cell_type":"markdown","metadata":{"id":"RYR1STLzZQzS"},"source":["L'objectif est simplement ici de tester le gel des poids pour les modèles de feature extraction (backbone) qui fournissent des descripteurs aux modèles de segmentation et ce afin de pouvoir comparer les différences de performances."]},{"cell_type":"code","execution_count":null,"metadata":{"id":"Jyxe-hvRZJ5C"},"outputs":[],"source":["def test_backbone_freezing(num_classes=19):\n"," \"\"\"\n"," Fonction de test pour vérifier que le gel du backbone fonctionne correctement.\n"," Version corrigée utilisant encoder_freeze.\n"," \"\"\"\n"," print(\"🧪 TEST DU GEL DU BACKBONE (VERSION CORRIGÉE)\")\n"," print(\"=\" * 60)\n","\n"," # === CONFIGURATIONS DE TEST ===\n"," test_configs = [\n"," {\n"," 'name': 'U-Net_mobilenetv2_frozen_test',\n"," 'architecture': 'U-Net',\n"," 'encoder': 'mobilenetv2',\n"," 'encoder_freeze': True, # GELÉ - devrait avoir beaucoup de params gelés\n"," 'params': {'learning_rate': 1e-4, 'img_size': (224, 224)},\n"," 'type': 'frozen_test'\n"," },\n"," {\n"," 'name': 'U-Net_mobilenetv2_trainable_test',\n"," 'architecture': 'U-Net',\n"," 'encoder': 'mobilenetv2',\n"," 'encoder_freeze': False, # DÉGELÉ - devrait avoir peu de params gelés\n"," 'params': {'learning_rate': 1e-4, 'img_size': (224, 224)},\n"," 'type': 'trainable_test'\n"," }\n"," ]\n","\n"," results = {}\n","\n"," # === EXÉCUTION DES TESTS ===\n"," for config in test_configs:\n"," print(f\"\\n🔍 Test: {config['name']}\")\n"," print(\"-\" * 50)\n","\n"," # Création du modèle avec verbose pour voir les détails\n"," model = create_model_from_config(config, num_classes, verbose=True)\n","\n"," # === COMPTAGE DÉTAILLÉ DES PARAMÈTRES ===\n"," total_params = model.count_params()\n"," trainable_params = sum([tf.keras.backend.count_params(w) for w in model.trainable_weights])\n"," frozen_params = total_params - trainable_params\n"," frozen_ratio = (frozen_params / total_params) * 100\n","\n"," print(f\"\\n📈 Résultat:\")\n"," print(f\" • Total: {total_params:,}\")\n"," print(f\" • Entraînables: {trainable_params:,}\")\n"," print(f\" • Gelés: {frozen_params:,}\")\n"," print(f\" • % gelés: {frozen_ratio:.1f}%\")\n","\n"," # === VALIDATION DU RÉSULTAT ===\n"," if config['encoder_freeze']:\n"," # Pour un modèle gelé, on s'attend à >15% de paramètres gelés\n"," expected_frozen_ratio = 20.0\n"," status = \"✅ OK\" if frozen_ratio > expected_frozen_ratio else \"❌ ÉCHEC\"\n"," print(f\" • Statut (gelé attendu >20%): {status}\")\n"," else:\n"," # Pour un modèle dégelé, on s'attend à <10% de paramètres gelés\n"," expected_frozen_ratio = 5\n"," status = \"✅ OK\" if frozen_ratio < expected_frozen_ratio else \"❌ ÉCHEC\"\n"," print(f\" • Statut (dégelé attendu <5%): {status}\")\n","\n"," # Stockage des résultats pour comparaison\n"," results[config['type']] = {\n"," 'total': total_params,\n"," 'trainable': trainable_params,\n"," 'frozen': frozen_params,\n"," 'frozen_ratio': frozen_ratio\n"," }\n","\n"," # === NETTOYAGE MÉMOIRE ===\n"," del model\n"," tf.keras.backend.clear_session()\n","\n"," # === RÉSUMÉ COMPARATIF ===\n"," print(f\"\\n🎯 RÉSUMÉ COMPARATIF\")\n"," print(\"=\" * 60)\n","\n"," if 'frozen_test' in results and 'trainable_test' in results:\n"," frozen_result = results['frozen_test']\n"," trainable_result = results['trainable_test']\n","\n"," print(f\"Modèle gelé (encoder_freeze=True):\")\n"," print(f\" • Params gelés: {frozen_result['frozen']:,} ({frozen_result['frozen_ratio']:.1f}%)\")\n","\n"," print(f\"Modèle dégelé (encoder_freeze=False):\")\n"," print(f\" • Params gelés: {trainable_result['frozen']:,} ({trainable_result['frozen_ratio']:.1f}%)\")\n","\n"," difference = frozen_result['frozen'] - trainable_result['frozen']\n"," print(f\"Différence: {difference:,} paramètres correctement gelés\")\n","\n"," if difference > 100000: # Au moins 100k params de différence\n"," print(\"✅ TEST GLOBAL: Le gel du backbone fonctionne correctement!\")\n"," else:\n"," print(\"❌ TEST GLOBAL: Le gel du backbone ne semble pas fonctionner!\")"]},{"cell_type":"markdown","metadata":{"id":"EvurXI0XZzng"},"source":["On va lister les différentes expérimentations et vérifier le nombre de poids gelés pour deux configurations différentes : une avec backbone entrainable, l'autre non."]},{"cell_type":"code","execution_count":null,"metadata":{"colab":{"base_uri":"https://localhost:8080/"},"executionInfo":{"elapsed":3128,"status":"ok","timestamp":1754919043370,"user":{"displayName":"Grégoire","userId":"01249298206117462766"},"user_tz":-120},"id":"AsgetKA1ZM0y","outputId":"9219fca1-e3c1-4a38-c906-8ea00ff7b1f5"},"outputs":[{"output_type":"stream","name":"stdout","text":["📊 PLAN D'EXPÉRIMENTATIONS - 17 configurations\n","======================================================================\n","🔧 Backbones utilisés: ['efficientnetb0', 'mobilenetv2', 'resnet34']\n","🏗️ Architectures: U-Net, FPN\n","\n","🔹 BASELINE (4 configs):\n"," • U-Net_vanilla_augFalse \n"," • U-Net_vanilla_augTrue \n"," • FPN_vanilla_augFalse \n"," • FPN_vanilla_augTrue \n","\n","🔹 FROZEN PRETRAINED (10 configs):\n"," • U-Net_mobilenetv2_frozen_augFalse (gelé)\n"," • U-Net_mobilenetv2_frozen_augTrue (gelé)\n"," • U-Net_efficientnetb0_frozen_augFalse (gelé)\n"," • U-Net_efficientnetb0_frozen_augTrue (gelé)\n"," • FPN_mobilenetv2_frozen_augFalse (gelé)\n"," • FPN_mobilenetv2_frozen_augTrue (gelé)\n"," • FPN_efficientnetb0_frozen_augFalse (gelé)\n"," • FPN_efficientnetb0_frozen_augTrue (gelé)\n"," • FPN_resnet34_frozen_augFalse (gelé)\n"," • FPN_resnet34_frozen_augTrue (gelé)\n","\n","🔹 FINETUNED PRETRAINED (3 configs):\n"," • U-Net_mobilenetv2_finetune_augTrue (dégelé)\n"," • FPN_efficientnetb0_finetune_augTrue (dégelé)\n"," • FPN_resnet34_finetune_augTrue (dégelé)\n","\n","======================================================================\n","LANCEMENT DU TEST DE VALIDATION\n","======================================================================\n","🧪 TEST DU GEL DU BACKBONE (VERSION CORRIGÉE)\n","============================================================\n","\n","🔍 Test: U-Net_mobilenetv2_frozen_test\n","--------------------------------------------------\n","🔧 Création du modèle 'U-Net_mobilenetv2_frozen_test'\n"," Architecture: U-Net\n"," Encoder: mobilenetv2\n"," Encoder gelé: True\n"," 📊 Paramètres: 8,050,051 total | 5,824,195 entraînables | 2,225,856 gelés\n","\n","📈 Résultat:\n"," • Total: 8,050,051\n"," • Entraînables: 5,824,195\n"," • Gelés: 2,225,856\n"," • % gelés: 27.7%\n"," • Statut (gelé attendu >20%): ✅ OK\n","\n","🔍 Test: U-Net_mobilenetv2_trainable_test\n","--------------------------------------------------\n","🔧 Création du modèle 'U-Net_mobilenetv2_trainable_test'\n"," Architecture: U-Net\n"," Encoder: mobilenetv2\n"," Encoder gelé: False\n"," 📊 Paramètres: 8,050,051 total | 8,013,955 entraînables | 36,096 gelés\n","\n","📈 Résultat:\n"," • Total: 8,050,051\n"," • Entraînables: 8,013,955\n"," • Gelés: 36,096\n"," • % gelés: 0.4%\n"," • Statut (dégelé attendu <5%): ✅ OK\n","\n","🎯 RÉSUMÉ COMPARATIF\n","============================================================\n","Modèle gelé (encoder_freeze=True):\n"," • Params gelés: 2,225,856 (27.7%)\n","Modèle dégelé (encoder_freeze=False):\n"," • Params gelés: 36,096 (0.4%)\n","Différence: 2,189,760 paramètres correctement gelés\n","✅ TEST GLOBAL: Le gel du backbone fonctionne correctement!\n"]}],"source":["# Affichage des configurations\n","print(f\"📊 PLAN D'EXPÉRIMENTATIONS - {len(experimental_configurations)} configurations\")\n","print(\"=\" * 70)\n","\n","# Affichage des backbones utilisés\n","used_backbones = set()\n","for config in experimental_configurations:\n"," if config['encoder']:\n"," used_backbones.add(config['encoder'])\n","\n","print(f\"🔧 Backbones utilisés: {sorted(used_backbones)}\")\n","print(f\"🏗️ Architectures: U-Net, FPN\")\n","\n","# Grouper par type pour un affichage plus clair\n","config_types = {}\n","for config in experimental_configurations:\n"," config_type = config['type']\n"," if config_type not in config_types:\n"," config_types[config_type] = []\n"," config_types[config_type].append(config)\n","\n","for type_name, configs in config_types.items():\n"," print(f\"\\n🔹 {type_name.upper().replace('_', ' ')} ({len(configs)} configs):\")\n"," for config in configs:\n"," freeze_status = \"gelé\" if config['encoder_freeze'] else \"dégelé\"\n"," encoder_info = f\"({freeze_status})\" if config['encoder'] else \"\"\n"," print(f\" • {config['name']} {encoder_info}\")\n","\n","# === LANCEMENT DU TEST ===\n","print(f\"\\n\" + \"=\"*70)\n","print(\"LANCEMENT DU TEST DE VALIDATION\")\n","print(\"=\"*70)\n","\n","# Décommentez pour exécuter le test\n","test_backbone_freezing(num_classes=19)"]},{"cell_type":"code","execution_count":null,"metadata":{"colab":{"base_uri":"https://localhost:8080/"},"executionInfo":{"elapsed":4429,"status":"ok","timestamp":1754919047801,"user":{"displayName":"Grégoire","userId":"01249298206117462766"},"user_tz":-120},"id":"9TWHaWHA6z66","outputId":"f367450c-779c-4243-d8a6-1e016ecbf199"},"outputs":[{"output_type":"stream","name":"stdout","text":["\n","======================================================================\n","🏗️ ANALYSE COMPLÈTE: U-Net_mobilenetv2_frozen_augFalse\n","======================================================================\n","📊 Configuration:\n"," • Architecture: U-Net\n"," • Backbone: mobilenetv2\n"," • Encoder gelé: ✅ Oui\n"," • Type: Frozen Pretrained\n"," • Augmentation: ❌ Non\n"," • Learning rate: 0.0001\n"," • Époques prévues: 20\n"," • Taille images: (224, 224)\n","\n","🔧 Création du modèle...\n","\n","🔢 Paramètres:\n"," • Total: 8,050,051\n"," • Entraînables: 5,824,195 (72.3%)\n"," • Gelés: 2,225,856 (27.7%)\n","\n","🏗️ AUTRES COUCHES (résumé):\n"," • 93 couches entraînables\n"," • 103 couches gelées\n"," • (Utilisez detailed_layers=True pour voir toutes les couches)\n","\n","🎯 Validation du gel du backbone:\n"," • Statut global: ✅ OK\n"," • Backbone correctement gelé (27.7% ≥ 20%)\n"," • Plage attendue pour mobilenetv2: 20-35%\n","\n","✅ Analyse terminée avec succès\n","\n","======================================================================\n","🏗️ ANALYSE COMPLÈTE: U-Net_mobilenetv2_frozen_augFalse\n","======================================================================\n","📊 Configuration:\n"," • Architecture: U-Net\n"," • Backbone: mobilenetv2\n"," • Encoder gelé: ✅ Oui\n"," • Type: Frozen Pretrained\n"," • Augmentation: ❌ Non\n"," • Learning rate: 0.0001\n"," • Époques prévues: 20\n"," • Taille images: (224, 224)\n","\n","🔧 Création du modèle...\n","\n","🔢 Paramètres:\n"," • Total: 8,050,051\n"," • Entraînables: 5,824,195 (72.3%)\n"," • Gelés: 2,225,856 (27.7%)\n","\n","🔍 TOUTES LES COUCHES (🟢 = Entraînable, 🔴 = Gelé):\n","----------------------------------------------------------------------\n"," 1. 🔴 input_layer 0 params - GELÉ\n"," 2. 🔴 Conv1_pad 0 params - GELÉ\n"," 3. 🔴 Conv1 864 params - GELÉ\n"," 4. 🟢 bn_Conv1 128 params - ENTRAÎNABLE\n"," 5. 🔴 Conv1_relu 0 params - GELÉ\n"," 6. 🔴 expanded_conv_depthwise 288 params - GELÉ\n"," 7. 🟢 expanded_conv_depthwise_BN 128 params - ENTRAÎNABLE\n"," 8. 🔴 expanded_conv_depthwise_relu 0 params - GELÉ\n"," 9. 🔴 expanded_conv_project 512 params - GELÉ\n","10. 🟢 expanded_conv_project_BN 64 params - ENTRAÎNABLE\n","11. 🔴 block_1_expand 1,536 params - GELÉ\n","12. 🟢 block_1_expand_BN 384 params - ENTRAÎNABLE\n","13. 🔴 block_1_expand_relu 0 params - GELÉ\n","14. 🔴 block_1_pad 0 params - GELÉ\n","15. 🔴 block_1_depthwise 864 params - GELÉ\n","16. 🟢 block_1_depthwise_BN 384 params - ENTRAÎNABLE\n","17. 🔴 block_1_depthwise_relu 0 params - GELÉ\n","18. 🔴 block_1_project 2,304 params - GELÉ\n","19. 🟢 block_1_project_BN 96 params - ENTRAÎNABLE\n","20. 🔴 block_2_expand 3,456 params - GELÉ\n","21. 🟢 block_2_expand_BN 576 params - ENTRAÎNABLE\n","22. 🔴 block_2_expand_relu 0 params - GELÉ\n","23. 🔴 block_2_depthwise 1,296 params - GELÉ\n","24. 🟢 block_2_depthwise_BN 576 params - ENTRAÎNABLE\n","25. 🔴 block_2_depthwise_relu 0 params - GELÉ\n","26. 🔴 block_2_project 3,456 params - GELÉ\n","27. 🟢 block_2_project_BN 96 params - ENTRAÎNABLE\n","28. 🔴 block_2_add 0 params - GELÉ\n","29. 🔴 block_3_expand 3,456 params - GELÉ\n","30. 🟢 block_3_expand_BN 576 params - ENTRAÎNABLE\n","31. 🔴 block_3_expand_relu 0 params - GELÉ\n","32. 🔴 block_3_pad 0 params - GELÉ\n","33. 🔴 block_3_depthwise 1,296 params - GELÉ\n","34. 🟢 block_3_depthwise_BN 576 params - ENTRAÎNABLE\n","35. 🔴 block_3_depthwise_relu 0 params - GELÉ\n","36. 🔴 block_3_project 4,608 params - GELÉ\n","37. 🟢 block_3_project_BN 128 params - ENTRAÎNABLE\n","38. 🔴 block_4_expand 6,144 params - GELÉ\n","39. 🟢 block_4_expand_BN 768 params - ENTRAÎNABLE\n","40. 🔴 block_4_expand_relu 0 params - GELÉ\n","41. 🔴 block_4_depthwise 1,728 params - GELÉ\n","42. 🟢 block_4_depthwise_BN 768 params - ENTRAÎNABLE\n","43. 🔴 block_4_depthwise_relu 0 params - GELÉ\n","44. 🔴 block_4_project 6,144 params - GELÉ\n","45. 🟢 block_4_project_BN 128 params - ENTRAÎNABLE\n","46. 🔴 block_4_add 0 params - GELÉ\n","47. 🔴 block_5_expand 6,144 params - GELÉ\n","48. 🟢 block_5_expand_BN 768 params - ENTRAÎNABLE\n","49. 🔴 block_5_expand_relu 0 params - GELÉ\n","50. 🔴 block_5_depthwise 1,728 params - GELÉ\n","51. 🟢 block_5_depthwise_BN 768 params - ENTRAÎNABLE\n","52. 🔴 block_5_depthwise_relu 0 params - GELÉ\n","53. 🔴 block_5_project 6,144 params - GELÉ\n","54. 🟢 block_5_project_BN 128 params - ENTRAÎNABLE\n","55. 🔴 block_5_add 0 params - GELÉ\n","56. 🔴 block_6_expand 6,144 params - GELÉ\n","57. 🟢 block_6_expand_BN 768 params - ENTRAÎNABLE\n","58. 🔴 block_6_expand_relu 0 params - GELÉ\n","59. 🔴 block_6_pad 0 params - GELÉ\n","60. 🔴 block_6_depthwise 1,728 params - GELÉ\n","61. 🟢 block_6_depthwise_BN 768 params - ENTRAÎNABLE\n","62. 🔴 block_6_depthwise_relu 0 params - GELÉ\n","63. 🔴 block_6_project 12,288 params - GELÉ\n","64. 🟢 block_6_project_BN 256 params - ENTRAÎNABLE\n","65. 🔴 block_7_expand 24,576 params - GELÉ\n","66. 🟢 block_7_expand_BN 1,536 params - ENTRAÎNABLE\n","67. 🔴 block_7_expand_relu 0 params - GELÉ\n","68. 🔴 block_7_depthwise 3,456 params - GELÉ\n","69. 🟢 block_7_depthwise_BN 1,536 params - ENTRAÎNABLE\n","70. 🔴 block_7_depthwise_relu 0 params - GELÉ\n","71. 🔴 block_7_project 24,576 params - GELÉ\n","72. 🟢 block_7_project_BN 256 params - ENTRAÎNABLE\n","73. 🔴 block_7_add 0 params - GELÉ\n","74. 🔴 block_8_expand 24,576 params - GELÉ\n","75. 🟢 block_8_expand_BN 1,536 params - ENTRAÎNABLE\n","76. 🔴 block_8_expand_relu 0 params - GELÉ\n","77. 🔴 block_8_depthwise 3,456 params - GELÉ\n","78. 🟢 block_8_depthwise_BN 1,536 params - ENTRAÎNABLE\n","79. 🔴 block_8_depthwise_relu 0 params - GELÉ\n","80. 🔴 block_8_project 24,576 params - GELÉ\n","81. 🟢 block_8_project_BN 256 params - ENTRAÎNABLE\n","82. 🔴 block_8_add 0 params - GELÉ\n","83. 🔴 block_9_expand 24,576 params - GELÉ\n","84. 🟢 block_9_expand_BN 1,536 params - ENTRAÎNABLE\n","85. 🔴 block_9_expand_relu 0 params - GELÉ\n","86. 🔴 block_9_depthwise 3,456 params - GELÉ\n","87. 🟢 block_9_depthwise_BN 1,536 params - ENTRAÎNABLE\n","88. 🔴 block_9_depthwise_relu 0 params - GELÉ\n","89. 🔴 block_9_project 24,576 params - GELÉ\n","90. 🟢 block_9_project_BN 256 params - ENTRAÎNABLE\n","91. 🔴 block_9_add 0 params - GELÉ\n","92. 🔴 block_10_expand 24,576 params - GELÉ\n","93. 🟢 block_10_expand_BN 1,536 params - ENTRAÎNABLE\n","94. 🔴 block_10_expand_relu 0 params - GELÉ\n","95. 🔴 block_10_depthwise 3,456 params - GELÉ\n","96. 🟢 block_10_depthwise_BN 1,536 params - ENTRAÎNABLE\n","97. 🔴 block_10_depthwise_relu 0 params - GELÉ\n","98. 🔴 block_10_project 36,864 params - GELÉ\n","99. 🟢 block_10_project_BN 384 params - ENTRAÎNABLE\n","100. 🔴 block_11_expand 55,296 params - GELÉ\n","101. 🟢 block_11_expand_BN 2,304 params - ENTRAÎNABLE\n","102. 🔴 block_11_expand_relu 0 params - GELÉ\n","103. 🔴 block_11_depthwise 5,184 params - GELÉ\n","104. 🟢 block_11_depthwise_BN 2,304 params - ENTRAÎNABLE\n","105. 🔴 block_11_depthwise_relu 0 params - GELÉ\n","106. 🔴 block_11_project 55,296 params - GELÉ\n","107. 🟢 block_11_project_BN 384 params - ENTRAÎNABLE\n","108. 🔴 block_11_add 0 params - GELÉ\n","109. 🔴 block_12_expand 55,296 params - GELÉ\n","110. 🟢 block_12_expand_BN 2,304 params - ENTRAÎNABLE\n","111. 🔴 block_12_expand_relu 0 params - GELÉ\n","112. 🔴 block_12_depthwise 5,184 params - GELÉ\n","113. 🟢 block_12_depthwise_BN 2,304 params - ENTRAÎNABLE\n","114. 🔴 block_12_depthwise_relu 0 params - GELÉ\n","115. 🔴 block_12_project 55,296 params - GELÉ\n","116. 🟢 block_12_project_BN 384 params - ENTRAÎNABLE\n","117. 🔴 block_12_add 0 params - GELÉ\n","118. 🔴 block_13_expand 55,296 params - GELÉ\n","119. 🟢 block_13_expand_BN 2,304 params - ENTRAÎNABLE\n","120. 🔴 block_13_expand_relu 0 params - GELÉ\n","121. 🔴 block_13_pad 0 params - GELÉ\n","122. 🔴 block_13_depthwise 5,184 params - GELÉ\n","123. 🟢 block_13_depthwise_BN 2,304 params - ENTRAÎNABLE\n","124. 🔴 block_13_depthwise_relu 0 params - GELÉ\n","125. 🔴 block_13_project 92,160 params - GELÉ\n","126. 🟢 block_13_project_BN 640 params - ENTRAÎNABLE\n","127. 🔴 block_14_expand 153,600 params - GELÉ\n","128. 🟢 block_14_expand_BN 3,840 params - ENTRAÎNABLE\n","129. 🔴 block_14_expand_relu 0 params - GELÉ\n","130. 🔴 block_14_depthwise 8,640 params - GELÉ\n","131. 🟢 block_14_depthwise_BN 3,840 params - ENTRAÎNABLE\n","132. 🔴 block_14_depthwise_relu 0 params - GELÉ\n","133. 🔴 block_14_project 153,600 params - GELÉ\n","134. 🟢 block_14_project_BN 640 params - ENTRAÎNABLE\n","135. 🔴 block_14_add 0 params - GELÉ\n","136. 🔴 block_15_expand 153,600 params - GELÉ\n","137. 🟢 block_15_expand_BN 3,840 params - ENTRAÎNABLE\n","138. 🔴 block_15_expand_relu 0 params - GELÉ\n","139. 🔴 block_15_depthwise 8,640 params - GELÉ\n","140. 🟢 block_15_depthwise_BN 3,840 params - ENTRAÎNABLE\n","141. 🔴 block_15_depthwise_relu 0 params - GELÉ\n","142. 🔴 block_15_project 153,600 params - GELÉ\n","143. 🟢 block_15_project_BN 640 params - ENTRAÎNABLE\n","144. 🔴 block_15_add 0 params - GELÉ\n","145. 🔴 block_16_expand 153,600 params - GELÉ\n","146. 🟢 block_16_expand_BN 3,840 params - ENTRAÎNABLE\n","147. 🔴 block_16_expand_relu 0 params - GELÉ\n","148. 🔴 block_16_depthwise 8,640 params - GELÉ\n","149. 🟢 block_16_depthwise_BN 3,840 params - ENTRAÎNABLE\n","150. 🔴 block_16_depthwise_relu 0 params - GELÉ\n","151. 🔴 block_16_project 307,200 params - GELÉ\n","152. 🟢 block_16_project_BN 1,280 params - ENTRAÎNABLE\n","153. 🔴 Conv_1 409,600 params - GELÉ\n","154. 🟢 Conv_1_bn 5,120 params - ENTRAÎNABLE\n","155. 🔴 out_relu 0 params - GELÉ\n","156. 🟢 decoder_stage0_upsampling 0 params - ENTRAÎNABLE (DECODER)\n","157. 🟢 decoder_stage0_concat 0 params - ENTRAÎNABLE (DECODER)\n","158. 🟢 decoder_stage0a_conv 4,276,224 params - ENTRAÎNABLE (DECODER)\n","159. 🟢 decoder_stage0a_bn 1,024 params - ENTRAÎNABLE (DECODER)\n","160. 🟢 decoder_stage0a_relu 0 params - ENTRAÎNABLE (DECODER)\n","161. 🟢 decoder_stage0b_conv 589,824 params - ENTRAÎNABLE (DECODER)\n","162. 🟢 decoder_stage0b_bn 1,024 params - ENTRAÎNABLE (DECODER)\n","163. 🟢 decoder_stage0b_relu 0 params - ENTRAÎNABLE (DECODER)\n","164. 🟢 decoder_stage1_upsampling 0 params - ENTRAÎNABLE (DECODER)\n","165. 🟢 decoder_stage1_concat 0 params - ENTRAÎNABLE (DECODER)\n","166. 🟢 decoder_stage1a_conv 516,096 params - ENTRAÎNABLE (DECODER)\n","167. 🟢 decoder_stage1a_bn 512 params - ENTRAÎNABLE (DECODER)\n","168. 🟢 decoder_stage1a_relu 0 params - ENTRAÎNABLE (DECODER)\n","169. 🟢 decoder_stage1b_conv 147,456 params - ENTRAÎNABLE (DECODER)\n","170. 🟢 decoder_stage1b_bn 512 params - ENTRAÎNABLE (DECODER)\n","171. 🟢 decoder_stage1b_relu 0 params - ENTRAÎNABLE (DECODER)\n","172. 🟢 decoder_stage2_upsampling 0 params - ENTRAÎNABLE (DECODER)\n","173. 🟢 decoder_stage2_concat 0 params - ENTRAÎNABLE (DECODER)\n","174. 🟢 decoder_stage2a_conv 156,672 params - ENTRAÎNABLE (DECODER)\n","175. 🟢 decoder_stage2a_bn 256 params - ENTRAÎNABLE (DECODER)\n","176. 🟢 decoder_stage2a_relu 0 params - ENTRAÎNABLE (DECODER)\n","177. 🟢 decoder_stage2b_conv 36,864 params - ENTRAÎNABLE (DECODER)\n","178. 🟢 decoder_stage2b_bn 256 params - ENTRAÎNABLE (DECODER)\n","179. 🟢 decoder_stage2b_relu 0 params - ENTRAÎNABLE (DECODER)\n","180. 🟢 decoder_stage3_upsampling 0 params - ENTRAÎNABLE (DECODER)\n","181. 🟢 decoder_stage3_concat 0 params - ENTRAÎNABLE (DECODER)\n","182. 🟢 decoder_stage3a_conv 46,080 params - ENTRAÎNABLE (DECODER)\n","183. 🟢 decoder_stage3a_bn 128 params - ENTRAÎNABLE (DECODER)\n","184. 🟢 decoder_stage3a_relu 0 params - ENTRAÎNABLE (DECODER)\n","185. 🟢 decoder_stage3b_conv 9,216 params - ENTRAÎNABLE (DECODER)\n","186. 🟢 decoder_stage3b_bn 128 params - ENTRAÎNABLE (DECODER)\n","187. 🟢 decoder_stage3b_relu 0 params - ENTRAÎNABLE (DECODER)\n","188. 🟢 decoder_stage4_upsampling 0 params - ENTRAÎNABLE (DECODER)\n","189. 🟢 decoder_stage4a_conv 4,608 params - ENTRAÎNABLE (DECODER)\n","190. 🟢 decoder_stage4a_bn 64 params - ENTRAÎNABLE (DECODER)\n","191. 🟢 decoder_stage4a_relu 0 params - ENTRAÎNABLE (DECODER)\n","192. 🟢 decoder_stage4b_conv 2,304 params - ENTRAÎNABLE (DECODER)\n","193. 🟢 decoder_stage4b_bn 64 params - ENTRAÎNABLE (DECODER)\n","194. 🟢 decoder_stage4b_relu 0 params - ENTRAÎNABLE (DECODER)\n","195. 🟢 final_conv 2,755 params - ENTRAÎNABLE (CLASSIFICATION)\n","196. 🟢 softmax 0 params - ENTRAÎNABLE\n","\n","🎯 Validation du gel du backbone:\n"," • Statut global: ✅ OK\n"," • Backbone correctement gelé (27.7% ≥ 20%)\n"," • Plage attendue pour mobilenetv2: 20-35%\n","\n","✅ Analyse terminée avec succès\n","\n","🔍 COUCHES - U-Net_mobilenetv2_frozen_augFalse\n","======================================================================\n","🟢 ENTRAÎNABLE | 🔴 GELÉ\n","----------------------------------------------------------------------\n"," 1. 🔴 input_layer 0 AUTRE\n"," 2. 🔴 Conv1_pad 0 AUTRE\n"," 3. 🔴 Conv1 864 AUTRE\n"," 4. 🟢 bn_Conv1 128 AUTRE\n"," 5. 🔴 Conv1_relu 0 AUTRE\n"," 6. 🔴 expanded_conv_depthwise 288 AUTRE\n"," 7. 🟢 expanded_conv_depthwise_BN 128 AUTRE\n"," 8. 🔴 expanded_conv_depthwise_relu 0 AUTRE\n"," 9. 🔴 expanded_conv_project 512 AUTRE\n","10. 🟢 expanded_conv_project_BN 64 AUTRE\n","11. 🔴 block_1_expand 1,536 AUTRE\n","12. 🟢 block_1_expand_BN 384 AUTRE\n","13. 🔴 block_1_expand_relu 0 AUTRE\n","14. 🔴 block_1_pad 0 AUTRE\n","15. 🔴 block_1_depthwise 864 AUTRE\n","16. 🟢 block_1_depthwise_BN 384 AUTRE\n","17. 🔴 block_1_depthwise_relu 0 AUTRE\n","18. 🔴 block_1_project 2,304 AUTRE\n","19. 🟢 block_1_project_BN 96 AUTRE\n","20. 🔴 block_2_expand 3,456 AUTRE\n","21. 🟢 block_2_expand_BN 576 AUTRE\n","22. 🔴 block_2_expand_relu 0 AUTRE\n","23. 🔴 block_2_depthwise 1,296 AUTRE\n","24. 🟢 block_2_depthwise_BN 576 AUTRE\n","25. 🔴 block_2_depthwise_relu 0 AUTRE\n","26. 🔴 block_2_project 3,456 AUTRE\n","27. 🟢 block_2_project_BN 96 AUTRE\n","28. 🔴 block_2_add 0 AUTRE\n","29. 🔴 block_3_expand 3,456 AUTRE\n","30. 🟢 block_3_expand_BN 576 AUTRE\n","31. 🔴 block_3_expand_relu 0 AUTRE\n","32. 🔴 block_3_pad 0 AUTRE\n","33. 🔴 block_3_depthwise 1,296 AUTRE\n","34. 🟢 block_3_depthwise_BN 576 AUTRE\n","35. 🔴 block_3_depthwise_relu 0 AUTRE\n","36. 🔴 block_3_project 4,608 AUTRE\n","37. 🟢 block_3_project_BN 128 AUTRE\n","38. 🔴 block_4_expand 6,144 AUTRE\n","39. 🟢 block_4_expand_BN 768 AUTRE\n","40. 🔴 block_4_expand_relu 0 AUTRE\n","41. 🔴 block_4_depthwise 1,728 AUTRE\n","42. 🟢 block_4_depthwise_BN 768 AUTRE\n","43. 🔴 block_4_depthwise_relu 0 AUTRE\n","44. 🔴 block_4_project 6,144 AUTRE\n","45. 🟢 block_4_project_BN 128 AUTRE\n","46. 🔴 block_4_add 0 AUTRE\n","47. 🔴 block_5_expand 6,144 AUTRE\n","48. 🟢 block_5_expand_BN 768 AUTRE\n","49. 🔴 block_5_expand_relu 0 AUTRE\n","50. 🔴 block_5_depthwise 1,728 AUTRE\n","51. 🟢 block_5_depthwise_BN 768 AUTRE\n","52. 🔴 block_5_depthwise_relu 0 AUTRE\n","53. 🔴 block_5_project 6,144 AUTRE\n","54. 🟢 block_5_project_BN 128 AUTRE\n","55. 🔴 block_5_add 0 AUTRE\n","56. 🔴 block_6_expand 6,144 AUTRE\n","57. 🟢 block_6_expand_BN 768 AUTRE\n","58. 🔴 block_6_expand_relu 0 AUTRE\n","59. 🔴 block_6_pad 0 AUTRE\n","60. 🔴 block_6_depthwise 1,728 AUTRE\n","61. 🟢 block_6_depthwise_BN 768 AUTRE\n","62. 🔴 block_6_depthwise_relu 0 AUTRE\n","63. 🔴 block_6_project 12,288 AUTRE\n","64. 🟢 block_6_project_BN 256 AUTRE\n","65. 🔴 block_7_expand 24,576 AUTRE\n","66. 🟢 block_7_expand_BN 1,536 AUTRE\n","67. 🔴 block_7_expand_relu 0 AUTRE\n","68. 🔴 block_7_depthwise 3,456 AUTRE\n","69. 🟢 block_7_depthwise_BN 1,536 AUTRE\n","70. 🔴 block_7_depthwise_relu 0 AUTRE\n","71. 🔴 block_7_project 24,576 AUTRE\n","72. 🟢 block_7_project_BN 256 AUTRE\n","73. 🔴 block_7_add 0 AUTRE\n","74. 🔴 block_8_expand 24,576 AUTRE\n","75. 🟢 block_8_expand_BN 1,536 AUTRE\n","76. 🔴 block_8_expand_relu 0 AUTRE\n","77. 🔴 block_8_depthwise 3,456 AUTRE\n","78. 🟢 block_8_depthwise_BN 1,536 AUTRE\n","79. 🔴 block_8_depthwise_relu 0 AUTRE\n","80. 🔴 block_8_project 24,576 AUTRE\n","81. 🟢 block_8_project_BN 256 AUTRE\n","82. 🔴 block_8_add 0 AUTRE\n","83. 🔴 block_9_expand 24,576 AUTRE\n","84. 🟢 block_9_expand_BN 1,536 AUTRE\n","85. 🔴 block_9_expand_relu 0 AUTRE\n","86. 🔴 block_9_depthwise 3,456 AUTRE\n","87. 🟢 block_9_depthwise_BN 1,536 AUTRE\n","88. 🔴 block_9_depthwise_relu 0 AUTRE\n","89. 🔴 block_9_project 24,576 AUTRE\n","90. 🟢 block_9_project_BN 256 AUTRE\n","91. 🔴 block_9_add 0 AUTRE\n","92. 🔴 block_10_expand 24,576 AUTRE\n","93. 🟢 block_10_expand_BN 1,536 AUTRE\n","94. 🔴 block_10_expand_relu 0 AUTRE\n","95. 🔴 block_10_depthwise 3,456 AUTRE\n","96. 🟢 block_10_depthwise_BN 1,536 AUTRE\n","97. 🔴 block_10_depthwise_relu 0 AUTRE\n","98. 🔴 block_10_project 36,864 AUTRE\n","99. 🟢 block_10_project_BN 384 AUTRE\n","100. 🔴 block_11_expand 55,296 AUTRE\n","101. 🟢 block_11_expand_BN 2,304 AUTRE\n","102. 🔴 block_11_expand_relu 0 AUTRE\n","103. 🔴 block_11_depthwise 5,184 AUTRE\n","104. 🟢 block_11_depthwise_BN 2,304 AUTRE\n","105. 🔴 block_11_depthwise_relu 0 AUTRE\n","106. 🔴 block_11_project 55,296 AUTRE\n","107. 🟢 block_11_project_BN 384 AUTRE\n","108. 🔴 block_11_add 0 AUTRE\n","109. 🔴 block_12_expand 55,296 AUTRE\n","110. 🟢 block_12_expand_BN 2,304 AUTRE\n","111. 🔴 block_12_expand_relu 0 AUTRE\n","112. 🔴 block_12_depthwise 5,184 AUTRE\n","113. 🟢 block_12_depthwise_BN 2,304 AUTRE\n","114. 🔴 block_12_depthwise_relu 0 AUTRE\n","115. 🔴 block_12_project 55,296 AUTRE\n","116. 🟢 block_12_project_BN 384 AUTRE\n","117. 🔴 block_12_add 0 AUTRE\n","118. 🔴 block_13_expand 55,296 AUTRE\n","119. 🟢 block_13_expand_BN 2,304 AUTRE\n","120. 🔴 block_13_expand_relu 0 AUTRE\n","121. 🔴 block_13_pad 0 AUTRE\n","122. 🔴 block_13_depthwise 5,184 AUTRE\n","123. 🟢 block_13_depthwise_BN 2,304 AUTRE\n","124. 🔴 block_13_depthwise_relu 0 AUTRE\n","125. 🔴 block_13_project 92,160 AUTRE\n","126. 🟢 block_13_project_BN 640 AUTRE\n","127. 🔴 block_14_expand 153,600 AUTRE\n","128. 🟢 block_14_expand_BN 3,840 AUTRE\n","129. 🔴 block_14_expand_relu 0 AUTRE\n","130. 🔴 block_14_depthwise 8,640 AUTRE\n","131. 🟢 block_14_depthwise_BN 3,840 AUTRE\n","132. 🔴 block_14_depthwise_relu 0 AUTRE\n","133. 🔴 block_14_project 153,600 AUTRE\n","134. 🟢 block_14_project_BN 640 AUTRE\n","135. 🔴 block_14_add 0 AUTRE\n","136. 🔴 block_15_expand 153,600 AUTRE\n","137. 🟢 block_15_expand_BN 3,840 AUTRE\n","138. 🔴 block_15_expand_relu 0 AUTRE\n","139. 🔴 block_15_depthwise 8,640 AUTRE\n","140. 🟢 block_15_depthwise_BN 3,840 AUTRE\n","141. 🔴 block_15_depthwise_relu 0 AUTRE\n","142. 🔴 block_15_project 153,600 AUTRE\n","143. 🟢 block_15_project_BN 640 AUTRE\n","144. 🔴 block_15_add 0 AUTRE\n","145. 🔴 block_16_expand 153,600 AUTRE\n","146. 🟢 block_16_expand_BN 3,840 AUTRE\n","147. 🔴 block_16_expand_relu 0 AUTRE\n","148. 🔴 block_16_depthwise 8,640 AUTRE\n","149. 🟢 block_16_depthwise_BN 3,840 AUTRE\n","150. 🔴 block_16_depthwise_relu 0 AUTRE\n","151. 🔴 block_16_project 307,200 AUTRE\n","152. 🟢 block_16_project_BN 1,280 AUTRE\n","153. 🔴 Conv_1 409,600 AUTRE\n","154. 🟢 Conv_1_bn 5,120 AUTRE\n","155. 🔴 out_relu 0 AUTRE\n","156. 🟢 decoder_stage0_upsampling 0 DECODER\n","157. 🟢 decoder_stage0_concat 0 DECODER\n","158. 🟢 decoder_stage0a_conv 4,276,224 DECODER\n","159. 🟢 decoder_stage0a_bn 1,024 DECODER\n","160. 🟢 decoder_stage0a_relu 0 DECODER\n","161. 🟢 decoder_stage0b_conv 589,824 DECODER\n","162. 🟢 decoder_stage0b_bn 1,024 DECODER\n","163. 🟢 decoder_stage0b_relu 0 DECODER\n","164. 🟢 decoder_stage1_upsampling 0 DECODER\n","165. 🟢 decoder_stage1_concat 0 DECODER\n","166. 🟢 decoder_stage1a_conv 516,096 DECODER\n","167. 🟢 decoder_stage1a_bn 512 DECODER\n","168. 🟢 decoder_stage1a_relu 0 DECODER\n","169. 🟢 decoder_stage1b_conv 147,456 DECODER\n","170. 🟢 decoder_stage1b_bn 512 DECODER\n","171. 🟢 decoder_stage1b_relu 0 DECODER\n","172. 🟢 decoder_stage2_upsampling 0 DECODER\n","173. 🟢 decoder_stage2_concat 0 DECODER\n","174. 🟢 decoder_stage2a_conv 156,672 DECODER\n","175. 🟢 decoder_stage2a_bn 256 DECODER\n","176. 🟢 decoder_stage2a_relu 0 DECODER\n","177. 🟢 decoder_stage2b_conv 36,864 DECODER\n","178. 🟢 decoder_stage2b_bn 256 DECODER\n","179. 🟢 decoder_stage2b_relu 0 DECODER\n","180. 🟢 decoder_stage3_upsampling 0 DECODER\n","181. 🟢 decoder_stage3_concat 0 DECODER\n","182. 🟢 decoder_stage3a_conv 46,080 DECODER\n","183. 🟢 decoder_stage3a_bn 128 DECODER\n","184. 🟢 decoder_stage3a_relu 0 DECODER\n","185. 🟢 decoder_stage3b_conv 9,216 DECODER\n","186. 🟢 decoder_stage3b_bn 128 DECODER\n","187. 🟢 decoder_stage3b_relu 0 DECODER\n","188. 🟢 decoder_stage4_upsampling 0 DECODER\n","189. 🟢 decoder_stage4a_conv 4,608 DECODER\n","190. 🟢 decoder_stage4a_bn 64 DECODER\n","191. 🟢 decoder_stage4a_relu 0 DECODER\n","192. 🟢 decoder_stage4b_conv 2,304 DECODER\n","193. 🟢 decoder_stage4b_bn 64 DECODER\n","194. 🟢 decoder_stage4b_relu 0 DECODER\n","195. 🟢 final_conv 2,755 AUTRE\n","196. 🟢 softmax 0 AUTRE\n","----------------------------------------------------------------------\n","📊 RÉSUMÉ: 93 entraînables, 103 gelés\n"]}],"source":["# Analyse standard (backbone + résumé)\n","analyze_model_comprehensive(experimental_configurations[4], num_classes=19)\n","\n","# Analyse complète (toutes les couches)\n","analyze_model_comprehensive(experimental_configurations[4], num_classes=19, detailed_layers=True)\n","\n","# Vue compacte des couches uniquement\n","show_all_layers_status(experimental_configurations[4], num_classes=19)"]},{"cell_type":"markdown","metadata":{"id":"yGNRUQdxT4xL"},"source":["## Préparation des données"]},{"cell_type":"markdown","metadata":{"id":"e5mbV7CtEfoS"},"source":["Ici, les données vont être prétraitées et des datasets d'entrainement, de validation et de test vont être constitués.\n","Il est important de noter la correspondance terminologique entre le jeu de données officiel et les jeux de données que nous constituerons dans le cadre de cet entraînement :\\\n","• Split 'train' Cityscapes → Train (80%) + Validation (20%)\\\n","• Split 'val' Cityscapes → Test final\\\n","• Split 'test' Cityscapes → non utilisé\n","\n","En effet, le dossier \"test\" de Cityscapes ne contient pas de masque de vérité et sert à effectuer un benchmark de performances dans le cadre d'un concours Kaggel (résultats cachés, donc)."]},{"cell_type":"code","execution_count":null,"metadata":{"id":"wMEw3js7CNZE"},"outputs":[],"source":["def prepare_cityscapes_dataset(\n"," cityscapes_dir,\n"," split='train',\n"," validation_split=0.2,\n"," batch_size=8,\n"," img_size=(224, 224),\n"," encoder=None,\n"," id_to_group_mapping=None\n","):\n"," \"\"\"\n"," Prépare les datasets Cityscapes avec prétraitement adapté au backbone\n","\n"," Args:\n"," cityscapes_dir (str): Chemin vers le répertoire Cityscapes\n"," split (str): Split Cityscapes à utiliser ('train', 'val', 'test')\n"," validation_split (float): Si split='train', proportion à garder pour validation\n"," batch_size (int): Taille des batches\n"," img_size (tuple): Taille des images (H, W)\n"," encoder (str): Nom du backbone pour preprocessing ou None\n"," id_to_group_mapping (list ou np.ndarray): Mapping des IDs de classes ou None\n","\n"," Returns:\n"," Dataset(s): Si split='train' et validation_split > 0 → (train_dataset, val_dataset)\n"," Sinon → dataset unique\n"," \"\"\"\n"," if not os.path.exists(cityscapes_dir):\n"," raise ValueError(f\"Le répertoire Cityscapes n'existe pas: {cityscapes_dir}\")\n","\n"," images_dir = os.path.join(cityscapes_dir, 'leftImg8bit', split)\n"," masks_dir = os.path.join(cityscapes_dir, 'gtFine', split)\n","\n"," if not os.path.exists(images_dir) or not os.path.exists(masks_dir):\n"," raise ValueError(f\"Split Cityscapes '{split}' non trouvé dans {cityscapes_dir}\")\n","\n"," image_paths = []\n"," mask_paths = []\n","\n"," print(f\"🔍 Lecture du split Cityscapes '{split}'...\")\n","\n"," for city in os.listdir(images_dir):\n"," city_img_dir = os.path.join(images_dir, city)\n"," city_mask_dir = os.path.join(masks_dir, city)\n","\n"," if os.path.isdir(city_img_dir) and os.path.isdir(city_mask_dir):\n"," for file_name in os.listdir(city_img_dir):\n"," if file_name.endswith('_leftImg8bit.png'):\n"," img_path = os.path.join(city_img_dir, file_name)\n"," mask_name = file_name.replace('_leftImg8bit.png', '_gtFine_labelIds.png')\n"," mask_path = os.path.join(city_mask_dir, mask_name)\n"," if os.path.exists(img_path) and os.path.exists(mask_path):\n"," image_paths.append(img_path)\n"," mask_paths.append(mask_path)\n"," else:\n"," print(f\"⚠️ Fichier manquant pour {file_name}\")\n","\n"," print(f\"📊 Split Cityscapes '{split}': {len(image_paths)} images trouvées\")\n","\n"," if len(image_paths) == 0:\n"," raise ValueError(f\"Aucun fichier valide trouvé pour le split '{split}'!\")\n","\n"," if split == 'train' and validation_split > 0:\n"," print(f\"✂️ Découpage en Train/Validation ({int((1-validation_split)*100)}%/{int(validation_split*100)}%)\")\n","\n"," combined = list(zip(image_paths, mask_paths))\n"," random.seed(42)\n"," random.shuffle(combined)\n"," image_paths, mask_paths = zip(*combined)\n","\n"," total_samples = len(image_paths)\n"," val_samples = int(total_samples * validation_split)\n"," train_samples = total_samples - val_samples\n","\n"," train_img_paths = list(image_paths[:train_samples])\n"," train_mask_paths = list(mask_paths[:train_samples])\n"," val_img_paths = list(image_paths[train_samples:])\n"," val_mask_paths = list(mask_paths[train_samples:])\n","\n"," print(f\" └─ 📚 Train final: {len(train_img_paths)} images\")\n"," print(f\" └─ 📊 Validation final: {len(val_img_paths)} images\")\n"," else:\n"," usage = \"Test\" if split == 'val' else split.title()\n"," print(f\"📋 Utilisation complète comme dataset de {usage}\")\n","\n"," train_img_paths, train_mask_paths = image_paths, mask_paths\n"," val_img_paths, val_mask_paths = [], []\n","\n"," preprocessing_fn = get_preprocessing_fn(encoder)\n","\n"," def load_and_preprocess(img_path, mask_path):\n"," img = tf.io.read_file(img_path)\n"," img = tf.image.decode_png(img, channels=3)\n"," img = tf.image.resize(img, img_size, method='bilinear')\n"," img = tf.cast(img, tf.float32)\n"," img = preprocessing_fn(img)\n","\n"," mask = tf.io.read_file(mask_path)\n"," mask = tf.image.decode_png(mask, channels=1)\n"," mask = tf.image.resize(mask, img_size, method='nearest')\n"," mask = tf.cast(mask, tf.int32)\n","\n"," if id_to_group_mapping is not None:\n"," mask = map_mask_ids(mask, id_to_group_mapping)\n","\n"," mask = tf.squeeze(mask, axis=-1)\n"," return img, mask\n","\n"," def augment_data(image, mask):\n"," mask = tf.expand_dims(mask, axis=-1)\n","\n"," if tf.random.uniform(()) > 0.5:\n"," image = tf.image.flip_left_right(image)\n"," mask = tf.image.flip_left_right(mask)\n","\n"," image = tf.image.random_brightness(image, 0.1)\n"," image = tf.clip_by_value(image, 0, 1)\n","\n"," mask = tf.squeeze(mask, axis=-1)\n"," return image, mask\n","\n"," def create_dataset(image_paths, mask_paths, augment=False, shuffle_data=True):\n"," dataset = tf.data.Dataset.from_tensor_slices((image_paths, mask_paths))\n"," dataset = dataset.map(load_and_preprocess, num_parallel_calls=tf.data.AUTOTUNE)\n","\n"," if augment:\n"," dataset = dataset.map(augment_data, num_parallel_calls=tf.data.AUTOTUNE)\n","\n"," if shuffle_data:\n"," dataset = dataset.shuffle(buffer_size=1000)\n","\n"," dataset = dataset.batch(batch_size).prefetch(tf.data.AUTOTUNE)\n"," return dataset\n","\n"," if split == 'train' and validation_split > 0:\n"," train_dataset = create_dataset(train_img_paths, train_mask_paths, augment=True, shuffle_data=True)\n"," val_dataset = create_dataset(val_img_paths, val_mask_paths, augment=False, shuffle_data=False)\n"," return train_dataset, val_dataset\n"," else:\n"," dataset = create_dataset(train_img_paths, train_mask_paths, augment=False, shuffle_data=(split=='train'))\n"," return dataset\n"]},{"cell_type":"code","execution_count":null,"metadata":{"colab":{"base_uri":"https://localhost:8080/"},"executionInfo":{"elapsed":48688,"status":"ok","timestamp":1754919206313,"user":{"displayName":"Grégoire","userId":"01249298206117462766"},"user_tz":-120},"id":"VimLczDnCt7l","outputId":"0df294c8-4ff4-4df0-f614-e32476753e2d"},"outputs":[{"output_type":"stream","name":"stdout","text":["=== PRÉPARATION DES DATASETS CITYSCAPES ===\n","📊 Paramètres:\n"," • Taille image: (224, 224)\n"," • Taille batch: 8\n"," • Répertoire: /content/data/cityscapes_unzipped\n","\n","🗂️ Stratégie de découpage:\n"," • Split 'train' Cityscapes → Train (80%) + Validation (20%)\n"," • Split 'val' Cityscapes → Test final\n","\n","--- Découpage du split 'train' Cityscapes en Train/Validation ---\n","🔍 Lecture du split Cityscapes 'train'...\n","📊 Split Cityscapes 'train': 2975 images trouvées\n","✂️ Découpage en Train/Validation (80%/20%)\n"," └─ 📚 Train final: 2380 images\n"," └─ 📊 Validation final: 595 images\n","ℹ️ Modèle vanilla - pas de preprocessing\n","\n","--- Création du dataset de TEST à partir du split 'val' Cityscapes ---\n","🔍 Lecture du split Cityscapes 'val'...\n","📊 Split Cityscapes 'val': 500 images trouvées\n","📋 Utilisation complète comme dataset de Test\n","ℹ️ Modèle vanilla - pas de preprocessing\n","\n","✅ Datasets créés avec succès:\n"," 📚 Train: 2380 images (avec augmentation)\n"," 📊 Validation: 595 images (sans augmentation)\n"," 🧪 Test: 500 images (sans augmentation)\n","\n","📋 Récapitulatif:\n"," • Train dataset: pour l'entraînement des modèles\n"," • Validation dataset: pour la sélection des hyperparamètres\n"," • Test dataset: pour l'évaluation finale (données jamais vues)\n","\n","🔍 Vérification des formes:\n"," Train: images = (8, 224, 224, 3), masques = (8, 224, 224)\n"," Validation: images = (8, 224, 224, 3), masques = (8, 224, 224)\n"," Test: images = (8, 224, 224, 3), masques = (8, 224, 224)\n"]}],"source":["# === UTILISATION AVEC TERMINOLOGIE CLAIRE ===\n","# Définir les paramètres\n","BATCH_SIZE = 8\n","IMG_SIZE = (224, 224)\n","\n","# Vérifier si id_to_group existe, sinon définir None\n","try:\n"," id_mapping = id_to_group\n","except NameError:\n"," print(\"⚠️ Variable 'id_to_group' non définie, pas de mapping des IDs\")\n"," id_mapping = None\n","\n","print(\"=== PRÉPARATION DES DATASETS CITYSCAPES ===\")\n","print(f\"📊 Paramètres:\")\n","print(f\" • Taille image: {IMG_SIZE}\")\n","print(f\" • Taille batch: {BATCH_SIZE}\")\n","print(f\" • Répertoire: {cityscapes_dir}\")\n","\n","print(f\"\\n🗂️ Stratégie de découpage:\")\n","print(f\" • Split 'train' Cityscapes → Train (80%) + Validation (20%)\")\n","print(f\" • Split 'val' Cityscapes → Test final\")\n","\n","try:\n"," # 1. Créer TRAIN + VALIDATION à partir du split 'train' de Cityscapes\n"," print(f\"\\n--- Découpage du split 'train' Cityscapes en Train/Validation ---\")\n"," train_dataset, val_dataset = prepare_cityscapes_dataset(\n"," cityscapes_dir,\n"," split='train', # On utilise le split 'train' de Cityscapes\n"," batch_size=BATCH_SIZE,\n"," img_size=IMG_SIZE,\n"," validation_split=0.2, # 80% train, 20% validation\n"," id_to_group_mapping=id_mapping\n"," )\n","\n"," # 2. Créer TEST à partir du split 'val' de Cityscapes\n"," print(f\"\\n--- Création du dataset de TEST à partir du split 'val' Cityscapes ---\")\n"," test_dataset = prepare_cityscapes_dataset(\n"," cityscapes_dir,\n"," split='val', # On utilise le split 'val' de Cityscapes comme notre test\n"," batch_size=BATCH_SIZE,\n"," img_size=IMG_SIZE,\n"," id_to_group_mapping=id_mapping\n"," )\n","\n"," print(f\"\\n✅ Datasets créés avec succès:\")\n"," print(f\" 📚 Train: {2380} images (avec augmentation)\")\n"," print(f\" 📊 Validation: {595} images (sans augmentation)\")\n"," print(f\" 🧪 Test: {500} images (sans augmentation)\")\n","\n"," print(f\"\\n📋 Récapitulatif:\")\n"," print(f\" • Train dataset: pour l'entraînement des modèles\")\n"," print(f\" • Validation dataset: pour la sélection des hyperparamètres\")\n"," print(f\" • Test dataset: pour l'évaluation finale (données jamais vues)\")\n","\n"," # Vérification rapide\n"," print(f\"\\n🔍 Vérification des formes:\")\n"," for name, dataset in [(\"Train\", train_dataset), (\"Validation\", val_dataset), (\"Test\", test_dataset)]:\n"," sample = next(iter(dataset))\n"," print(f\" {name}: images = {sample[0].shape}, masques = {sample[1].shape}\")\n","\n","except Exception as e:\n"," print(f\"❌ Erreur: {e}\")\n"," print(\"Vérifications:\")\n"," print(\"1. Variable 'cityscapes_dir' définie ?\")\n"," print(\"2. Fichiers Cityscapes extraits ?\")\n"," print(\"3. Structure correcte (leftImg8bit/, gtFine/) ?\")\n"," raise"]},{"cell_type":"markdown","metadata":{"id":"q4W-fiAxSwWR"},"source":["## Visualisation des données"]},{"cell_type":"markdown","metadata":{"id":"aSz8Oa3sS3_n"},"source":["Une visualisation de quelques images (train, val, test) sera proposée avec les masques de vérité dont on propose une colorisation tab10 puis une colorisation plus spécifique"]},{"cell_type":"code","execution_count":null,"metadata":{"colab":{"base_uri":"https://localhost:8080/","height":1000,"output_embedded_package_id":"1SBKflUkIHfCCqcNzpwoqmSKVChyGKj2Z"},"executionInfo":{"elapsed":52363,"status":"ok","timestamp":1754916545770,"user":{"displayName":"Grégoire","userId":"01249298206117462766"},"user_tz":-120},"id":"mWO3yOl3Oq12","outputId":"df51e8a0-719f-44f9-a87a-b172f488740f"},"outputs":[{"output_type":"display_data","data":{"text/plain":"Output hidden; open in https://colab.research.google.com to view."},"metadata":{}}],"source":["def visualize_cityscapes_samples(\n"," datasets_dict=None,\n"," cityscapes_dir=None,\n"," num_samples=3,\n"," save_dir=None,\n"," show_stats=True,\n"," show_legend=True\n","):\n"," \"\"\"\n"," Visualise des échantillons Cityscapes avec options flexibles\n","\n"," Args:\n"," datasets_dict: {'train': train_dataset, 'val': val_dataset, 'test': test_dataset}\n"," cityscapes_dir: Chemin vers données brutes (si datasets_dict=None)\n"," num_samples: Nombre d'échantillons par dataset\n"," save_dir: Répertoire de sauvegarde (optionnel)\n"," show_stats: Afficher statistiques des classes\n"," show_legend: Créer légende des classes\n"," \"\"\"\n","\n"," # Mode 1: Depuis les datasets TensorFlow (préféré)\n"," if datasets_dict is not None:\n"," print(\"📊 Visualisation depuis les datasets TensorFlow...\")\n"," _visualize_from_datasets(datasets_dict, num_samples, save_dir, show_stats)\n","\n"," # Mode 2: Depuis les fichiers bruts (fallback)\n"," elif cityscapes_dir is not None:\n"," print(\"📁 Visualisation depuis les fichiers bruts...\")\n"," _visualize_from_files(cityscapes_dir, num_samples, save_dir, show_stats)\n","\n"," else:\n"," raise ValueError(\"Fournir soit datasets_dict soit cityscapes_dir\")\n","\n"," # Créer la légende si demandée\n"," if show_legend:\n"," _create_legend(save_dir)\n","\n","def _visualize_from_datasets(datasets_dict, num_samples, save_dir, show_stats):\n"," \"\"\"Visualise depuis les datasets TensorFlow (après preprocessing)\"\"\"\n","\n"," for split_name, dataset in datasets_dict.items():\n"," print(f\"\\n--- {split_name.title()} Dataset ---\")\n","\n"," sample_count = 0\n"," for batch in dataset.take(num_samples):\n"," images, masks = batch\n","\n"," # Prendre le premier échantillon du batch\n"," image = images[0]\n"," mask = masks[0]\n","\n"," # Créer la visualisation\n"," fig = _create_sample_plot(image.numpy(), mask.numpy(),\n"," f\"{split_name.title()} Sample {sample_count+1}\")\n","\n"," # Sauvegarder si demandé\n"," if save_dir:\n"," os.makedirs(save_dir, exist_ok=True)\n"," fig_path = os.path.join(save_dir, f'{split_name}_sample_{sample_count+1}.png')\n"," fig.savefig(fig_path, dpi=150, bbox_inches='tight')\n"," print(f\"💾 Sauvegardé: {fig_path}\")\n","\n"," plt.show()\n"," plt.close(fig)\n","\n"," # Afficher les stats si demandé\n"," if show_stats:\n"," _show_class_stats(mask.numpy())\n","\n"," sample_count += 1\n"," if sample_count >= num_samples:\n"," break\n","\n","def _visualize_from_files(cityscapes_dir, num_samples, save_dir, show_stats):\n"," \"\"\"Visualise depuis les fichiers bruts (fallback)\"\"\"\n","\n"," for split in ['train', 'val']:\n"," print(f\"\\n--- {split.title()} (Raw Files) ---\")\n","\n"," # Collecter les chemins (réutilise la logique de prepare_cityscapes_dataset)\n"," images_dir = os.path.join(cityscapes_dir, 'leftImg8bit', split)\n"," masks_dir = os.path.join(cityscapes_dir, 'gtFine', split)\n","\n"," if not os.path.exists(images_dir):\n"," print(f\"⚠️ Split '{split}' non trouvé, ignoré\")\n"," continue\n","\n"," image_paths, mask_paths = _collect_sample_paths(images_dir, masks_dir, num_samples)\n","\n"," for i, (img_path, mask_path) in enumerate(zip(image_paths, mask_paths)):\n"," # Charger avec PIL\n"," img = np.array(PILImage.open(img_path).resize((512, 256), PILImage.Resampling.BILINEAR))\n"," mask_raw = np.array(PILImage.open(mask_path).resize((512, 256), PILImage.Resampling.NEAREST))\n","\n"," # Appliquer le mapping (vérifier que id_to_group existe)\n"," if 'id_to_group' in globals():\n"," mask = id_to_group[mask_raw]\n"," else:\n"," print(\"⚠️ Variable 'id_to_group' non trouvée, utilisation brute\")\n"," mask = mask_raw\n","\n"," # Normaliser l'image pour l'affichage\n"," img_normalized = img.astype(np.float32) / 255.0\n","\n"," # Créer la visualisation\n"," fig = _create_sample_plot(img_normalized, mask, f\"{split.title()} Raw Sample {i+1}\")\n","\n"," # Sauvegarder si demandé\n"," if save_dir:\n"," os.makedirs(save_dir, exist_ok=True)\n"," fig_path = os.path.join(save_dir, f'{split}_raw_sample_{i+1}.png')\n"," fig.savefig(fig_path, dpi=150, bbox_inches='tight')\n"," print(f\"💾 Sauvegardé: {fig_path}\")\n","\n"," plt.show()\n"," plt.close(fig)\n","\n"," # Afficher les stats si demandé\n"," if show_stats:\n"," _show_class_stats(mask)\n","\n","def _create_sample_plot(image, mask, title):\n"," \"\"\"Crée le plot avec les 3 vues (factorisation du code de visualisation)\"\"\"\n","\n"," fig, axes = plt.subplots(1, 3, figsize=(15, 5))\n"," fig.suptitle(title, fontsize=14)\n","\n"," # Vue 1: Image originale\n"," axes[0].imshow(image)\n"," axes[0].set_title(\"Image Originale\")\n"," axes[0].axis('off')\n","\n"," # Vue 2: Masque avec IDs (en tab10)\n"," im = axes[1].imshow(mask, cmap='tab10', vmin=0, vmax=mapping_config['num_classes']-1)\n"," axes[1].set_title(\"Masque (IDs de classes)\")\n"," axes[1].axis('off')\n","\n"," # Vue 3: Masque colorisé avec nos couleurs\n"," colored_mask = _colorize_mask(mask)\n"," axes[2].imshow(colored_mask)\n"," axes[2].set_title(\"Masque Segmenté (Coloré)\")\n"," axes[2].axis('off')\n","\n"," plt.tight_layout()\n"," return fig\n","\n","def _colorize_mask(mask):\n"," \"\"\"Applique la colorisation du masque (logique centralisée)\"\"\"\n"," colored_mask = np.zeros((mask.shape[0], mask.shape[1], 3), dtype=np.uint8)\n"," for group_id in range(mapping_config['num_classes']):\n"," colored_mask[mask == group_id] = mapping_config['group_colors'][group_id]\n"," return colored_mask\n","\n","def _show_class_stats(mask):\n"," \"\"\"Affiche les statistiques des classes (factorisation)\"\"\"\n"," unique_classes, counts = np.unique(mask, return_counts=True)\n"," print(\"\\n📊 Classes présentes dans cette image:\")\n"," for class_id, count in zip(unique_classes, counts):\n"," if class_id < len(mapping_config['group_names']): # Vérification bounds\n"," percentage = (count / mask.size) * 100\n"," print(f\" {mapping_config['group_names'][class_id]}: {count} pixels ({percentage:.1f}%)\")\n","\n","def _collect_sample_paths(images_dir, masks_dir, num_samples):\n"," \"\"\"Collecte aléatoirement des chemins d'images (factorisation)\"\"\"\n"," image_paths, mask_paths = [], []\n","\n"," for city in os.listdir(images_dir):\n"," city_img_dir = os.path.join(images_dir, city)\n"," city_mask_dir = os.path.join(masks_dir, city)\n","\n"," if os.path.isdir(city_img_dir) and os.path.isdir(city_mask_dir):\n"," for file_name in os.listdir(city_img_dir):\n"," if file_name.endswith('_leftImg8bit.png'):\n"," img_path = os.path.join(city_img_dir, file_name)\n"," mask_name = file_name.replace('_leftImg8bit.png', '_gtFine_labelIds.png')\n"," mask_path = os.path.join(city_mask_dir, mask_name)\n","\n"," if os.path.exists(mask_path):\n"," image_paths.append(img_path)\n"," mask_paths.append(mask_path)\n","\n"," # Sélection aléatoire\n"," if len(image_paths) > num_samples:\n"," selected_indices = np.random.choice(len(image_paths), size=num_samples, replace=False)\n"," image_paths = [image_paths[i] for i in selected_indices]\n"," mask_paths = [mask_paths[i] for i in selected_indices]\n","\n"," return image_paths, mask_paths\n","\n","def _create_legend(save_dir):\n"," \"\"\"Crée la légende des classes (factorisation)\"\"\"\n"," fig, ax = plt.subplots(figsize=(8, 5))\n"," fig.suptitle(\"Légende des classes de segmentation\", fontsize=14)\n","\n"," for i, (name, color) in enumerate(zip(mapping_config['group_names'], mapping_config['group_colors'])):\n"," ax.fill_between([0, 1], [i, i], [i+0.9, i+0.9], color=[c/255 for c in color])\n"," ax.text(1.1, i+0.5, name, va='center', fontsize=12)\n","\n"," ax.set_xlim(0, 2.5)\n"," ax.set_ylim(len(mapping_config['group_names'])-0.1, -0.1)\n"," ax.axis('off')\n","\n"," if save_dir:\n"," os.makedirs(save_dir, exist_ok=True)\n"," legend_path = os.path.join(save_dir, 'class_legend.png')\n"," fig.savefig(legend_path, dpi=150, bbox_inches='tight')\n"," print(f\"🏷️ Légende sauvegardée: {legend_path}\")\n","\n"," plt.show()\n"," plt.close(fig)\n","\n","# === UTILISATION SIMPLIFIÉE ===\n","print(\"=== VISUALISATION DES ÉCHANTILLONS CITYSCAPES ===\")\n","\n","# Méthode préférée: depuis les datasets TensorFlow\n","datasets_dict = {\n"," 'train': train_dataset,\n"," 'val': val_dataset,\n"," 'test': test_dataset\n","}\n","\n","# Créer le dossier de sauvegarde\n","viz_dir = os.path.join(data_dir, 'samples') if 'data_dir' in globals() else './visualizations'\n","\n","# Visualiser avec toutes les options\n","visualize_cityscapes_samples(\n"," datasets_dict=datasets_dict,\n"," num_samples=2, # 2 échantillons par dataset\n"," save_dir=viz_dir,\n"," show_stats=True,\n"," show_legend=True\n",")"]},{"cell_type":"markdown","metadata":{"id":"BcRXJIz6Ts6q"},"source":["## Entrainement du modèle"]},{"cell_type":"code","execution_count":15,"metadata":{"colab":{"base_uri":"https://localhost:8080/"},"executionInfo":{"elapsed":60,"status":"ok","timestamp":1754980948757,"user":{"displayName":"Grégoire","userId":"01249298206117462766"},"user_tz":-120},"id":"lUxAq504vmjH","outputId":"81e42c45-ca7d-4d73-c853-3f8d08331813"},"outputs":[{"output_type":"stream","name":"stdout","text":["🚀 MODE PRODUCTION ACTIVÉ\n","📊 Configuration globale: Device=/CPU:0, Max_Epochs=20\n"," Patience EarlyStopping=5, Patience LR=3\n"]}],"source":["# ====================================================================\n","# SCRIPT D'ENTRAÎNEMENT OPTIMISÉ POUR SEGMENTATION MULTI-CONFIGURATIONS\n","# AVEC SYSTÈME DE REPRISE AUTOMATIQUE\n","# ====================================================================\n","\n","# 🚨 PARAMÈTRE PRINCIPAL : Basculer entre TEST et PRODUCTION\n","TEST_MODE = False # Mettre False pour le mode production\n","\n","# Configuration selon le mode\n","if TEST_MODE:\n"," print(\"🧪 MODE TEST ACTIVÉ\")\n"," DEVICE_CONFIG = '/CPU:0'\n"," MAX_EPOCHS = 2 # Très rapide pour les tests\n"," EARLY_STOPPING_PATIENCE = 2 # Arrêt anticipé si pas d'amélioration pendant 2 époques\n"," LR_REDUCE_PATIENCE = 1 # Réduction LR si pas d'amélioration pendant 1 époque\n"," SUBSET_SIZE = 20\n"," # Taille fixe pour les tests (plus rapide)\n"," TEST_IMG_SIZE = (128, 128)\n"," TEST_BATCH_SIZE = 2\n"," TEST_MAX_CONFIGS = None # Nombre max de configs à tester\n"," # Expérience MLflow déjà définie en début de notebook\n","else:\n"," print(\"🚀 MODE PRODUCTION ACTIVÉ\")\n"," DEVICE_CONFIG = '/GPU:0' if tf.config.list_physical_devices('GPU') else '/CPU:0'\n"," MAX_EPOCHS = 20 # 20 époques max (beaucoup de modèles à tester)\n"," EARLY_STOPPING_PATIENCE = 5 # Arrêt anticipé si pas d'amélioration pendant 5 époques\n"," LR_REDUCE_PATIENCE = 3 # Réduction LR si pas d'amélioration pendant 3 époques\n"," SUBSET_SIZE = None\n"," # En production, utilise les paramètres de chaque configuration\n"," # Expérience MLflow déjà définie en début de notebook\n","\n","print(f\"📊 Configuration globale: Device={DEVICE_CONFIG}, Max_Epochs={MAX_EPOCHS}\")\n","print(f\" Patience EarlyStopping={EARLY_STOPPING_PATIENCE}, Patience LR={LR_REDUCE_PATIENCE}\")"]},{"cell_type":"code","execution_count":16,"metadata":{"id":"RqI-umcJvt9u","executionInfo":{"status":"ok","timestamp":1754980948757,"user_tz":-120,"elapsed":58,"user":{"displayName":"Grégoire","userId":"01249298206117462766"}}},"outputs":[],"source":["# ====================================================================\n","# SYSTÈME DE REPRISE AUTOMATIQUE - VÉRIFICATION MLFLOW\n","# ====================================================================\n","\n","def get_completed_experiments(experiment_name=\"OC Projet 8\"):\n"," \"\"\"\n"," Récupère la liste des expériences déjà terminées dans MLflow\n"," Retourne un set des noms de runs terminés\n"," \"\"\"\n"," try:\n"," print(f\"🔍 Vérification des expériences déjà terminées dans '{experiment_name}'...\")\n","\n"," # Récupérer l'expérience par nom\n"," experiment = mlflow.get_experiment_by_name(experiment_name)\n"," if experiment is None:\n"," print(f\"⚠️ Expérience '{experiment_name}' non trouvée\")\n"," return set()\n","\n"," # Rechercher tous les runs de cette expérience\n"," runs = mlflow.search_runs(\n"," experiment_ids=[experiment.experiment_id],\n"," filter_string=\"status = 'FINISHED'\", # Seulement les runs terminés avec succès\n"," output_format=\"pandas\"\n"," )\n","\n"," if runs.empty:\n"," print(\"📊 Aucune expérience terminée trouvée\")\n"," return set()\n","\n"," # Extraire les noms des runs terminés\n"," completed_runs = set()\n"," for idx, run in runs.iterrows():\n"," run_name = run.get('tags.mlflow.runName', '')\n"," if run_name:\n"," completed_runs.add(run_name)\n","\n"," print(f\"✅ {len(completed_runs)} expériences terminées trouvées:\")\n"," for run_name in sorted(completed_runs):\n"," print(f\" - {run_name}\")\n","\n"," return completed_runs\n","\n"," except Exception as e:\n"," print(f\"⚠️ Erreur lors de la vérification MLflow: {e}\")\n"," print(\"🔄 Continuera sans vérification (tous les modèles seront entraînés)\")\n"," return set()\n","\n","\n","def generate_stable_run_name(config, test_mode=False):\n"," \"\"\"\n"," Génère un nom de run stable (sans date) basé sur la configuration\n"," Format: {config_name}_{mode}\n"," \"\"\"\n"," base_name = config['name']\n"," mode_suffix = \"test\" if test_mode else \"prod\"\n"," return f\"{base_name}_{mode_suffix}\"\n","\n","\n","def filter_configs_to_run(experimental_configurations, experiment_name=\"OC Projet 8\"):\n"," \"\"\"\n"," Filtre les configurations pour ne garder que celles non encore entraînées\n"," \"\"\"\n"," completed_runs = get_completed_experiments(experiment_name)\n","\n"," if not completed_runs:\n"," print(\"🚀 Aucune expérience terminée trouvée, tous les modèles seront entraînés\")\n"," return list(range(len(experimental_configurations)))\n","\n"," configs_to_run = []\n"," skipped_configs = []\n","\n"," for idx, config in enumerate(experimental_configurations):\n"," expected_run_name = generate_stable_run_name(config, TEST_MODE)\n","\n"," if expected_run_name in completed_runs:\n"," skipped_configs.append((idx, config['name'], expected_run_name))\n"," else:\n"," configs_to_run.append(idx)\n","\n"," # Rapport de filtrage\n"," print(f\"\\n📊 FILTRAGE DES CONFIGURATIONS:\")\n"," print(f\"✅ À entraîner: {len(configs_to_run)} configurations\")\n"," print(f\"⏭️ Déjà terminées: {len(skipped_configs)} configurations\")\n","\n"," if skipped_configs:\n"," print(f\"\\n⏭️ CONFIGURATIONS IGNORÉES (déjà terminées):\")\n"," for idx, config_name, run_name in skipped_configs:\n"," print(f\" [{idx}] {config_name} → {run_name}\")\n","\n"," if configs_to_run:\n"," print(f\"\\n🚀 CONFIGURATIONS À ENTRAÎNER:\")\n"," for idx in configs_to_run:\n"," config_name = experimental_configurations[idx]['name']\n"," expected_run_name = generate_stable_run_name(experimental_configurations[idx], TEST_MODE)\n"," print(f\" [{idx}] {config_name} → {expected_run_name}\")\n","\n"," return configs_to_run"]},{"cell_type":"code","execution_count":17,"metadata":{"id":"yXkQoZssvwTt","executionInfo":{"status":"ok","timestamp":1754980948757,"user_tz":-120,"elapsed":18,"user":{"displayName":"Grégoire","userId":"01249298206117462766"}}},"outputs":[],"source":["# ====================================================================\n","# CALLBACK MLFLOW PERSONNALISÉ POUR LOGGING DÉTAILLÉ\n","# ====================================================================\n","\n","class ExperimentMLflowCallback(tf.keras.callbacks.Callback):\n"," \"\"\"Callback MLflow personnalisé pour logging complet des métriques\"\"\"\n","\n"," def __init__(self, config, run_id, batch_size, img_size):\n"," super().__init__()\n"," self.config = config\n"," self.run_id = run_id\n"," self.batch_size = batch_size\n"," self.img_size = img_size\n","\n"," def on_epoch_end(self, epoch, logs=None):\n"," \"\"\"Log des métriques à chaque époque (sans doublons)\"\"\"\n"," logs = logs or {}\n"," for name, value in logs.items():\n"," # Éviter les doublons : ignorer les métriques \"validation_\"\n"," if not name.startswith('validation_'):\n"," mlflow.log_metric(name, float(value), step=epoch)\n","\n"," def on_train_begin(self, logs=None):\n"," \"\"\"Log des paramètres au début de l'entraînement\"\"\"\n"," # Log config du modèle\n"," mlflow.log_param(\"model_architecture\", self.config[\"architecture\"])\n"," mlflow.log_param(\"model_name\", self.config[\"name\"])\n"," mlflow.log_param(\"encoder\", self.config.get(\"encoder\", \"None\"))\n"," mlflow.log_param(\"encoder_freeze\", self.config.get(\"encoder_freeze\", False))\n","\n"," # Log config du script (utilise les paramètres passés au callback)\n"," mlflow.log_param(\"script_batch_size\", self.batch_size)\n"," mlflow.log_param(\"script_img_size\", str(self.img_size))\n"," mlflow.log_param(\"test_mode\", TEST_MODE)\n"," mlflow.log_param(\"max_epochs\", MAX_EPOCHS)\n"," mlflow.log_param(\"device\", DEVICE_CONFIG)\n"," mlflow.log_param(\"subset_size\", SUBSET_SIZE)\n","\n","\n","class KerasModelCheckpoint(tf.keras.callbacks.Callback):\n"," \"\"\"Callback personnalisé pour sauvegarder en format .keras natif\"\"\"\n","\n"," def __init__(self, filepath, monitor='val_loss', save_best_only=True, mode='auto', verbose=1):\n"," super().__init__()\n"," self.filepath = filepath\n"," self.monitor = monitor\n"," self.save_best_only = save_best_only\n"," self.verbose = verbose\n","\n"," if mode == 'auto':\n"," if 'acc' in monitor or monitor.startswith('fmeasure'):\n"," self.mode = 'max'\n"," else:\n"," self.mode = 'min'\n"," else:\n"," self.mode = mode\n","\n"," if self.mode == 'min':\n"," self.best = float('inf')\n"," else:\n"," self.best = -float('inf')\n","\n"," def on_epoch_end(self, epoch, logs=None):\n"," logs = logs or {}\n"," current = logs.get(self.monitor)\n","\n"," if current is None:\n"," if self.verbose > 0:\n"," print(f'Warning: Metric {self.monitor} not found in logs.')\n"," return\n","\n"," if self.save_best_only:\n"," if (self.mode == 'min' and current < self.best) or (self.mode == 'max' and current > self.best):\n"," self.best = current\n","\n"," if self.verbose > 0:\n"," print(f'Epoch {epoch + 1}: {self.monitor} improved from {self.best} to {current:.5f}, saving model to {self.filepath}')\n","\n"," # SAUVEGARDE NATIVE KERAS - AUCUN WARNING !\n"," self.model.save(self.filepath, save_format='keras')\n"," else:\n"," if self.verbose > 0:\n"," print(f'Epoch {epoch + 1}: {self.monitor} did not improve from {self.best:.5f}')\n"," else:\n"," if self.verbose > 0:\n"," print(f'Epoch {epoch + 1}: saving model to {self.filepath}')\n"," self.model.save(self.filepath, save_format='keras')"]},{"cell_type":"code","source":["def calculate_detailed_metrics(model, test_dataset, num_classes, class_names=None):\n"," \"\"\"\n"," Calcule uniquement l'IoU par classe et les statistiques détaillées\n"," (Le mean_iou global est maintenant calculé par la métrique officielle)\n"," \"\"\"\n"," print(\" 📊 Analyse détaillée par classe...\")\n","\n"," all_true_masks = []\n"," all_pred_masks = []\n","\n"," # Collecter toutes les prédictions\n"," for batch_idx, (images, true_masks) in enumerate(test_dataset):\n"," predictions = model.predict(images, verbose=0)\n"," pred_masks = tf.argmax(predictions, axis=-1)\n","\n"," all_true_masks.extend(true_masks.numpy())\n"," all_pred_masks.extend(pred_masks.numpy())\n","\n"," if batch_idx % 10 == 0:\n"," print(f\" Batch {batch_idx + 1} traité...\")\n","\n"," all_true_masks = np.array(all_true_masks)\n"," all_pred_masks = np.array(all_pred_masks)\n","\n"," print(f\" Total d'images évaluées: {len(all_true_masks)}\")\n","\n"," # Calcul IoU par classe (pour analyse détaillée)\n"," class_ious = []\n"," class_stats = []\n","\n"," for class_id in range(num_classes):\n"," # Masques binaires pour cette classe\n"," true_binary = (all_true_masks == class_id)\n"," pred_binary = (all_pred_masks == class_id)\n","\n"," # Calcul IoU\n"," intersection = np.logical_and(true_binary, pred_binary).sum()\n"," union = np.logical_or(true_binary, pred_binary).sum()\n","\n"," iou = intersection / union if union > 0 else 0.0\n"," class_ious.append(iou)\n","\n"," # Statistiques de classe\n"," true_pixels = true_binary.sum()\n"," pred_pixels = pred_binary.sum()\n"," total_pixels = all_true_masks.size\n","\n"," class_name = class_names[class_id] if class_names else f\"Class_{class_id}\"\n","\n"," class_stats.append({\n"," 'class_id': class_id,\n"," 'class_name': class_name,\n"," 'iou': iou,\n"," 'true_pixels': true_pixels,\n"," 'pred_pixels': pred_pixels,\n"," 'true_percentage': 100 * true_pixels / total_pixels,\n"," 'pred_percentage': 100 * pred_pixels / total_pixels\n"," })\n","\n"," # Mean IoU calculé ici (pour cohérence avec l'affichage)\n"," mean_iou = np.mean(class_ious)\n","\n"," # Affichage des résultats détaillés\n"," print(f\"\\n📊 ANALYSE DÉTAILLÉE PAR CLASSE:\")\n"," print(\"-\" * 70)\n"," print(f\"{'Classe':<15} {'IoU':<8} {'Vrais %':<10} {'Prédits %':<12}\")\n"," print(\"-\" * 70)\n","\n"," for stats in class_stats:\n"," print(f\"{stats['class_name']:<15} {stats['iou']:<8.4f} {stats['true_percentage']:<10.2f} {stats['pred_percentage']:<12.2f}\")\n","\n"," print(\"-\" * 70)\n"," print(f\"{'MEAN IoU (calc)':<15} {mean_iou:<8.4f}\")\n"," print(\"💡 Le Mean IoU officiel (métrique Keras) fait référence\")\n","\n"," return {\n"," 'class_ious': class_ious,\n"," 'mean_iou': mean_iou, # Pour compatibilité avec le code existant\n"," 'class_stats': class_stats,\n"," 'all_true_masks': all_true_masks,\n"," 'all_pred_masks': all_pred_masks\n"," }\n"],"metadata":{"id":"9-rX20362Pz2","executionInfo":{"status":"ok","timestamp":1754980948758,"user_tz":-120,"elapsed":18,"user":{"displayName":"Grégoire","userId":"01249298206117462766"}}},"execution_count":18,"outputs":[]},{"cell_type":"code","source":["def setup_training_components(config, run_id):\n"," \"\"\"\n"," Setup complet : paramètres, datasets, modèle, compilation, callbacks\n"," VERSION CORRIGÉE avec métriques officielles Keras\n"," \"\"\"\n"," # === EXTRACTION DES PARAMÈTRES DE LA CONFIG ===\n"," if TEST_MODE:\n"," batch_size = TEST_BATCH_SIZE\n"," img_size = TEST_IMG_SIZE\n"," print(f\"🧪 Mode test: batch_size={batch_size}, img_size={img_size}\")\n"," else:\n"," batch_size = config['params']['batch_size']\n"," img_size = config['params']['img_size']\n"," print(f\"🚀 Mode production: batch_size={batch_size}, img_size={img_size}\")\n","\n"," # === PRÉPARATION DES DONNÉES ===\n"," print(\"\\n📊 Préparation des datasets...\")\n"," try:\n"," id_mapping = id_to_group\n"," mlflow.log_param(\"class_mapping\", \"custom_id_to_group\")\n"," except NameError:\n"," print(\"⚠️ Variable 'id_to_group' non définie\")\n"," id_mapping = None\n"," mlflow.log_param(\"class_mapping\", \"none\")\n","\n"," # Créer les datasets avec les bons paramètres\n"," train_dataset, val_dataset = prepare_cityscapes_dataset(\n"," cityscapes_dir,\n"," split='train',\n"," batch_size=batch_size,\n"," img_size=img_size,\n"," validation_split=0.2,\n"," id_to_group_mapping=id_mapping\n"," )\n"," test_dataset = prepare_cityscapes_dataset(\n"," cityscapes_dir,\n"," split='val',\n"," batch_size=batch_size,\n"," img_size=img_size,\n"," id_to_group_mapping=id_mapping\n"," )\n","\n"," if TEST_MODE and SUBSET_SIZE:\n"," train_batches = SUBSET_SIZE // batch_size\n"," val_batches = SUBSET_SIZE // (batch_size * 4)\n"," test_batches = SUBSET_SIZE // (batch_size * 4)\n"," train_samples_used = train_batches * batch_size\n"," val_samples_used = val_batches * batch_size\n"," test_samples_used = test_batches * batch_size\n","\n"," print(f\"🧪 Mode test: limitation appliquée\")\n"," print(f\" └─ 📚 Train utilisé: {train_samples_used} images ({train_batches} batches)\")\n"," print(f\" └─ 📊 Validation utilisé: {val_samples_used} images ({val_batches} batches)\")\n"," print(f\" └─ 🧪 Test utilisé: {test_samples_used} images ({test_batches} batches)\")\n","\n"," train_dataset = train_dataset.take(train_batches)\n"," val_dataset = val_dataset.take(val_batches)\n"," test_dataset = test_dataset.take(test_batches)\n","\n"," # === CRÉATION DU MODÈLE ===\n"," print(f\"\\n🏗️ Création du modèle {config['name']}...\")\n"," if TEST_MODE:\n"," config_modified = config.copy()\n"," config_modified['params'] = config['params'].copy()\n"," config_modified['params']['img_size'] = img_size\n"," config_modified['params']['batch_size'] = batch_size\n"," model_config = config_modified\n"," else:\n"," model_config = config\n","\n"," with tf.device(DEVICE_CONFIG):\n"," model = create_model_from_config(model_config, mapping_config['num_classes'])\n"," num_classes_from_model = model.output_shape[-1]\n"," print(f\"📊 Nombre de classes: {num_classes_from_model}\")\n"," total_params = model.count_params()\n"," trainable_params = sum([tf.keras.backend.count_params(w) for w in model.trainable_weights])\n"," print(f\"📊 Paramètres: {total_params:,} total | {trainable_params:,} entraînables\")\n","\n"," # === COMPILATION AVEC MÉTRIQUES OFFICIELLES KERAS ===\n"," print(\"\\n⚙️ Compilation du modèle avec métriques officielles...\")\n","\n"," # ✅ MÉTRIQUES OFFICIELLES KERAS - AUCUNE MÉTRIQUE PERSONNALISÉE !\n"," metrics_list = [\n"," # Accuracy officielle pour classification sparse\n"," tf.keras.metrics.SparseCategoricalAccuracy(name='sparse_categorical_accuracy'),\n","\n"," # IoU officielle Keras - LA métrique de référence pour segmentation\n"," tf.keras.metrics.MeanIoU(\n"," num_classes=num_classes_from_model,\n"," name='mean_iou',\n"," sparse_y_true=True, # Les labels sont des entiers (pas one-hot)\n"," sparse_y_pred=False # Les prédictions sont des probabilités\n"," )\n"," ]\n","\n"," # Métrique principale pour les callbacks\n"," primary_metric = 'val_mean_iou'\n"," primary_mode = 'max'\n","\n"," print(\"🎯 MÉTRIQUES OFFICIELLES KERAS POUR SEGMENTATION SÉMANTIQUE:\")\n"," print(\" 🔥 LOSS: sparse_categorical_crossentropy\")\n"," print(\" └─ Optimise la classification pixel-par-pixel\")\n"," print(\" 📊 MÉTRIQUES (100% officielles Keras):\")\n"," print(\" ├─ SparseCategoricalAccuracy: % pixels correctement classifiés\")\n"," print(\" └─ MeanIoU: Intersection over Union (métrique gold standard)\")\n"," print(\" 🎛️ CALLBACKS:\")\n"," print(f\" └─ Monitor: {primary_metric} (IoU = métrique de référence)\")\n"," print(\"✅ Configuration 100% standard et fiable !\")\n","\n"," # Compilation du modèle\n"," model.compile(\n"," optimizer=tf.keras.optimizers.Adam(learning_rate=0.001),\n"," loss='sparse_categorical_crossentropy',\n"," metrics=metrics_list\n"," )\n","\n"," # === CALLBACKS OPTIMISÉS ===\n"," print(f\"\\n📋 Configuration des callbacks...\")\n"," print(\"🎛️ CALLBACKS BASÉS SUR MEAN_IOU OFFICIEL:\")\n"," print(f\" ├─ EarlyStopping: Arrêt si {primary_metric} stagne {EARLY_STOPPING_PATIENCE} époques\")\n"," print(f\" ├─ ReduceLROnPlateau: Réduction LR si {primary_metric} stagne {LR_REDUCE_PATIENCE} époques\")\n"," print(f\" └─ ModelCheckpoint: Sauvegarde du meilleur modèle selon {primary_metric}\")\n","\n"," artifacts_dir = f\"artifacts/run_{run_id}\"\n"," os.makedirs(artifacts_dir, exist_ok=True)\n","\n"," mlflow_callback = ExperimentMLflowCallback(config, run_id, batch_size, img_size)\n","\n"," callbacks_list = [\n"," mlflow_callback,\n"," tf.keras.callbacks.ReduceLROnPlateau(\n"," monitor=primary_metric,\n"," factor=0.2,\n"," patience=LR_REDUCE_PATIENCE,\n"," verbose=1,\n"," min_lr=1e-6,\n"," mode=primary_mode\n"," ),\n"," tf.keras.callbacks.EarlyStopping(\n"," monitor=primary_metric,\n"," patience=EARLY_STOPPING_PATIENCE,\n"," restore_best_weights=True,\n"," verbose=1,\n"," mode=primary_mode\n"," ),\n"," KerasModelCheckpoint(\n"," filepath=os.path.join(artifacts_dir, 'best_model.keras'),\n"," monitor=primary_metric,\n"," save_best_only=True,\n"," mode=primary_mode,\n"," verbose=1\n"," ),\n"," tf.keras.callbacks.TensorBoard(\n"," log_dir=os.path.join(artifacts_dir, 'tensorboard'),\n"," histogram_freq=1\n"," )\n"," ]\n","\n"," return {\n"," 'model': model,\n"," 'train_dataset': train_dataset,\n"," 'val_dataset': val_dataset,\n"," 'test_dataset': test_dataset,\n"," 'callbacks_list': callbacks_list,\n"," 'artifacts_dir': artifacts_dir,\n"," 'num_classes_from_model': num_classes_from_model,\n"," 'batch_size': batch_size,\n"," 'img_size': img_size\n"," }\n","\n","def train_model_core(model, train_dataset, val_dataset, callbacks_list, max_epochs):\n"," \"\"\"\n"," Fonction pure d'entraînement - ne fait que l'entraînement\n"," \"\"\"\n"," print(f\"\\n🚀 Début de l'entraînement ({max_epochs} epochs max)...\")\n"," start_time = time.time()\n","\n"," with tf.device(DEVICE_CONFIG):\n"," history = model.fit(\n"," train_dataset,\n"," validation_data=val_dataset,\n"," epochs=max_epochs,\n"," callbacks=callbacks_list,\n"," verbose=1\n"," )\n","\n"," training_time = time.time() - start_time\n"," mlflow.log_metric(\"training_time_seconds\", training_time)\n"," print(f\"⏱️ Temps d'entraînement: {training_time:.2f}s\")\n","\n"," return history, training_time\n","\n","def evaluate_model_comprehensive(model, test_dataset, num_classes_from_model, artifacts_dir):\n"," \"\"\"\n"," Évaluation complète avec métriques officielles Keras + analyses détaillées\n"," \"\"\"\n"," # === ÉVALUATION AVEC MÉTRIQUES OFFICIELLES (même config que l'entraînement) ===\n"," print(\"\\n📊 Évaluation avec métriques officielles Keras...\")\n","\n"," # Créer exactement les mêmes métriques que pendant l'entraînement\n"," official_metrics = [\n"," tf.keras.metrics.SparseCategoricalAccuracy(name='sparse_categorical_accuracy'),\n"," tf.keras.metrics.MeanIoU(\n"," num_classes=num_classes_from_model,\n"," name='mean_iou',\n"," sparse_y_true=True,\n"," sparse_y_pred=False\n"," )\n"," ]\n","\n"," # Compiler temporairement avec les métriques officielles\n"," # (ça ne change pas les poids, juste les métriques)\n"," model.compile(\n"," optimizer=model.optimizer, # Garder l'optimizer existant\n"," loss=model.loss, # Garder la loss existante\n"," metrics=official_metrics\n"," )\n","\n"," # Évaluation officielle\n"," test_results = model.evaluate(test_dataset, verbose=1)\n","\n"," basic_test_metrics = {\n"," 'loss': test_results[0],\n"," 'sparse_categorical_accuracy': test_results[1],\n"," 'mean_iou': test_results[2]\n"," }\n","\n"," # Log dans MLflow\n"," for metric_name, value in basic_test_metrics.items():\n"," mlflow.log_metric(f\"test_{metric_name}\", value)\n"," print(f\"✅ Test {metric_name} (officiel): {value:.4f}\")\n","\n"," # === ÉVALUATION DÉTAILLÉE ADDITIONNELLE (pour analyse par classe) ===\n"," try:\n"," class_names = mapping_config['group_names'] if 'group_names' in mapping_config else None\n"," except:\n"," class_names = [f\"Class_{i}\" for i in range(num_classes_from_model)]\n","\n"," print(\"\\n🔍 Calcul des métriques détaillées par classe...\")\n"," detailed_metrics = calculate_detailed_metrics(\n"," model, test_dataset, num_classes_from_model, class_names\n"," )\n","\n"," # Log des métriques par classe (en complément)\n"," for i, iou in enumerate(detailed_metrics['class_ious']):\n"," mlflow.log_metric(f\"class_{i}_iou\", iou)\n","\n"," # === MATRICE DE CONFUSION ===\n"," print(\"\\n🔥 Génération de la matrice de confusion...\")\n"," confusion_matrix_data = generate_confusion_matrix(\n"," model, test_dataset, num_classes_from_model, class_names, artifacts_dir\n"," )\n","\n"," return basic_test_metrics, detailed_metrics, confusion_matrix_data, class_names\n","\n","def generate_confusion_matrix(model, test_dataset, num_classes, class_names, artifacts_dir):\n"," \"\"\"\n"," Génère la matrice de confusion sur le jeu de test et la sauvegarde (SANS AFFICHAGE)\n"," \"\"\"\n"," print(\"📊 Calcul de la matrice de confusion...\")\n","\n"," # Collecter toutes les prédictions\n"," y_true_all = []\n"," y_pred_all = []\n","\n"," for batch_images, batch_masks in test_dataset:\n"," # Prédictions du modèle\n"," predictions = model.predict(batch_images, verbose=0)\n"," pred_classes = tf.argmax(predictions, axis=-1)\n","\n"," # Aplatir les masques et prédictions\n"," true_flat = tf.reshape(batch_masks, [-1]).numpy()\n"," pred_flat = tf.reshape(pred_classes, [-1]).numpy()\n","\n"," y_true_all.extend(true_flat)\n"," y_pred_all.extend(pred_flat)\n","\n"," # Calcul de la matrice de confusion\n"," from sklearn.metrics import confusion_matrix\n"," import itertools\n","\n"," cm = confusion_matrix(y_true_all, y_pred_all, labels=range(num_classes))\n","\n"," # === CRÉATION DE LA VISUALISATION (SANS AFFICHAGE) ===\n"," plt.figure(figsize=(10, 8))\n"," plt.imshow(cm, interpolation='nearest', cmap=plt.cm.Blues)\n"," plt.title('Matrice de Confusion - Jeu de Test', fontsize=14)\n"," plt.colorbar()\n","\n"," # Définir les labels des axes\n"," if class_names:\n"," tick_marks = np.arange(len(class_names))\n"," plt.xticks(tick_marks, class_names, rotation=45, ha='right')\n"," plt.yticks(tick_marks, class_names)\n"," else:\n"," tick_marks = np.arange(num_classes)\n"," plt.xticks(tick_marks, [f\"Class_{i}\" for i in range(num_classes)])\n"," plt.yticks(tick_marks, [f\"Class_{i}\" for i in range(num_classes)])\n","\n"," # === AFFICHER LES VALEURS DANS CHAQUE CELLULE ===\n"," thresh = cm.max() / 2.0\n"," for i, j in itertools.product(range(cm.shape[0]), range(cm.shape[1])):\n"," plt.text(j, i, format(cm[i, j], 'd'),\n"," horizontalalignment=\"center\",\n"," verticalalignment=\"center\",\n"," color=\"white\" if cm[i, j] > thresh else \"black\",\n"," fontsize=10,\n"," fontweight='bold')\n","\n"," plt.ylabel('Vraie classe', fontsize=12)\n"," plt.xlabel('Classe prédite', fontsize=12)\n"," plt.tight_layout()\n","\n"," # Sauvegarder l'image (SANS AFFICHAGE)\n"," cm_image_path = os.path.join(artifacts_dir, 'confusion_matrix.png')\n"," plt.savefig(cm_image_path, dpi=300, bbox_inches='tight')\n"," plt.close() # ✅ FERMER LE PLOT - PAS D'AFFICHAGE\n","\n"," # Données JSON pour l'artefact\n"," confusion_matrix_data = {\n"," 'matrix': cm.tolist(), # Conversion numpy -> liste Python\n"," 'class_names': class_names if class_names else [f\"Class_{i}\" for i in range(num_classes)],\n"," 'total_samples': len(y_true_all),\n"," 'accuracy': float(np.trace(cm) / np.sum(cm)) # Accuracy globale\n"," }\n","\n"," print(f\"✅ Matrice de confusion sauvegardée : {cm_image_path}\")\n","\n"," return confusion_matrix_data\n","\n","def generate_training_visualizations(history, detailed_metrics, class_names, run_id, config, artifacts_dir):\n"," \"\"\"\n"," Génère toutes les visualisations : courbes d'entraînement + IoU par classe\n"," \"\"\"\n"," print(\"\\n📈 Création des visualisations...\")\n"," fig, axes = plt.subplots(2, 2, figsize=(15, 10))\n"," fig.suptitle(f'Entraînement {config[\"name\"]} - Run {run_id[:8]}', fontsize=16)\n","\n"," # Loss\n"," axes[0, 0].plot(history.history['loss'], label='Train Loss')\n"," axes[0, 0].plot(history.history['val_loss'], label='Val Loss')\n"," axes[0, 0].set_title('Loss')\n"," axes[0, 0].legend()\n"," axes[0, 0].grid(True)\n","\n"," # Accuracy\n"," axes[0, 1].plot(history.history['sparse_categorical_accuracy'], label='Train Acc')\n"," axes[0, 1].plot(history.history['val_sparse_categorical_accuracy'], label='Val Acc')\n"," axes[0, 1].set_title('Accuracy')\n"," axes[0, 1].legend()\n"," axes[0, 1].grid(True)\n","\n"," # Mean IoU\n"," axes[1, 0].plot(history.history['mean_iou'], label='Train Mean IoU')\n"," axes[1, 0].plot(history.history['val_mean_iou'], label='Val Mean IoU')\n"," axes[1, 0].set_title('Mean IoU')\n"," axes[1, 0].legend()\n"," axes[1, 0].grid(True)\n","\n"," # IoU par classe (bar chart)\n"," if class_names and len(class_names) == len(detailed_metrics['class_ious']):\n"," axes[1, 1].bar(range(len(detailed_metrics['class_ious'])), detailed_metrics['class_ious'])\n"," axes[1, 1].set_title(f'IoU par classe sur jeu de test (Mean: {detailed_metrics[\"mean_iou\"]:.3f})')\n"," axes[1, 1].set_xticks(range(len(class_names)))\n"," axes[1, 1].set_xticklabels(class_names, rotation=45)\n"," axes[1, 1].grid(True, axis='y')\n","\n"," plt.tight_layout()\n"," training_curves_path = os.path.join(artifacts_dir, 'training_curves.png')\n"," plt.savefig(training_curves_path, dpi=300, bbox_inches='tight')\n"," plt.close()\n","\n","def save_training_artifacts(history, config, detailed_metrics, confusion_matrix_data, artifacts_dir):\n"," \"\"\"\n"," Sauvegarde tous les artefacts : history, config, métriques, matrice de confusion\n"," \"\"\"\n"," print(\"\\n💾 Sauvegarde des artefacts...\")\n","\n"," # Histoire d'entraînement\n"," history_path = os.path.join(artifacts_dir, 'training_history.json')\n"," with open(history_path, 'w') as f:\n"," history_json = {k: [float(v) for v in values] for k, values in history.history.items()}\n"," json.dump(history_json, f, indent=2)\n","\n"," # Configuration du modèle\n"," config_path = os.path.join(artifacts_dir, 'model_config.json')\n"," with open(config_path, 'w') as f:\n"," json.dump(config, f, indent=2)\n","\n"," # Métriques détaillées - CONVERSION NUMPY VERS PYTHON POUR JSON\n"," detailed_path = os.path.join(artifacts_dir, 'detailed_metrics.json')\n"," with open(detailed_path, 'w') as f:\n"," class_stats_json = []\n"," for stats in detailed_metrics['class_stats']:\n"," stats_converted = {\n"," 'class_id': int(stats['class_id']),\n"," 'class_name': str(stats['class_name']),\n"," 'iou': float(stats['iou']),\n"," 'true_pixels': int(stats['true_pixels']),\n"," 'pred_pixels': int(stats['pred_pixels']),\n"," 'true_percentage': float(stats['true_percentage']),\n"," 'pred_percentage': float(stats['pred_percentage'])\n"," }\n"," class_stats_json.append(stats_converted)\n","\n"," detailed_json = {\n"," 'mean_iou': float(detailed_metrics['mean_iou']),\n"," 'class_ious': [float(iou) for iou in detailed_metrics['class_ious']],\n"," 'class_stats': class_stats_json\n"," }\n"," json.dump(detailed_json, f, indent=2)\n","\n"," # Matrice de confusion (si disponible)\n"," if confusion_matrix_data:\n"," confusion_path = os.path.join(artifacts_dir, 'confusion_matrix.json')\n"," with open(confusion_path, 'w') as f:\n"," json.dump(confusion_matrix_data, f, indent=2)\n","\n"," mlflow.log_artifacts(artifacts_dir)\n","\n","def train_segmentation_model_optimized(config, run_name=None):\n"," \"\"\"\n"," Fonction principale orchestrant l'entraînement complet - VERSION REFACTORISÉE\n"," \"\"\"\n"," with mlflow.start_run(run_name=run_name) as run:\n"," run_id = run.info.run_id\n"," print(f\"🔬 MLflow Run ID: {run_id}\")\n"," print(f\"🏷️ MLflow Run Name: {run_name}\")\n","\n"," try:\n"," # === 1. SETUP COMPLET ===\n"," setup_components = setup_training_components(config, run_id)\n","\n"," # === 2. ENTRAÎNEMENT ===\n"," history, training_time = train_model_core(\n"," setup_components['model'],\n"," setup_components['train_dataset'],\n"," setup_components['val_dataset'],\n"," setup_components['callbacks_list'],\n"," MAX_EPOCHS\n"," )\n","\n"," # === 3. ÉVALUATION COMPLÈTE ===\n"," basic_test_metrics, detailed_metrics, confusion_matrix_data, class_names = evaluate_model_comprehensive(\n"," setup_components['model'],\n"," setup_components['test_dataset'],\n"," setup_components['num_classes_from_model'],\n"," setup_components['artifacts_dir']\n"," )\n","\n"," # === 4. VISUALISATIONS ===\n"," generate_training_visualizations(\n"," history, detailed_metrics, class_names,\n"," run_id, config, setup_components['artifacts_dir']\n"," )\n","\n"," # === 5. SAUVEGARDE DES ARTEFACTS ===\n"," save_training_artifacts(\n"," history, config, detailed_metrics,\n"," confusion_matrix_data, setup_components['artifacts_dir']\n"," )\n","\n"," # === 6. RÉSULTATS FINAUX ===\n"," best_val_acc = max(history.history['val_sparse_categorical_accuracy'])\n"," results = {\n"," 'run_id': run_id,\n"," 'config': config,\n"," 'training_time': training_time,\n"," 'final_epoch': len(history.history['loss']),\n"," 'best_val_accuracy': best_val_acc,\n"," 'test_metrics': basic_test_metrics,\n"," 'detailed_metrics': detailed_metrics,\n"," 'confusion_matrix': confusion_matrix_data,\n"," 'artifacts_dir': setup_components['artifacts_dir'],\n"," 'num_classes': setup_components['num_classes_from_model']\n"," }\n","\n"," print(f\"\\n✅ Entraînement terminé!\")\n"," print(f\"📊 Meilleure val_accuracy: {best_val_acc:.4f}\")\n"," print(f\"📊 Mean IoU: {detailed_metrics['mean_iou']:.4f}\")\n"," test_acc = basic_test_metrics.get('sparse_categorical_accuracy', 0)\n"," print(f\"🧪 Test accuracy: {test_acc:.4f}\")\n"," print(f\"📁 Artefacts: {setup_components['artifacts_dir']}\")\n"," print(f\"🔬 MLflow Run: {run_id}\")\n","\n"," return results\n","\n"," except Exception as e:\n"," mlflow.log_param(\"error\", str(e))\n"," print(f\"❌ Erreur pendant l'entraînement: {e}\")\n"," raise"],"metadata":{"id":"mWvf8oQC0UBC","executionInfo":{"status":"ok","timestamp":1754980948830,"user_tz":-120,"elapsed":65,"user":{"displayName":"Grégoire","userId":"01249298206117462766"}}},"execution_count":19,"outputs":[]},{"cell_type":"code","execution_count":20,"metadata":{"id":"ThxE2s76sTaj","executionInfo":{"status":"ok","timestamp":1754980948853,"user_tz":-120,"elapsed":22,"user":{"displayName":"Grégoire","userId":"01249298206117462766"}}},"outputs":[],"source":["def run_experiments_optimized(configs_to_test=None, experiment_name=\"OC Projet 8\"):\n"," \"\"\"\n"," Version optimisée avec système de reprise automatique + NETTOYAGE VRAM SIMPLIFIÉ\n"," \"\"\"\n","\n"," print(f\"\\n{'='*60}\")\n"," print(f\"🚀 DÉBUT DES EXPÉRIENCES - Mode: {'TEST' if TEST_MODE else 'PRODUCTION'}\")\n"," print(f\"🔬 Expérience MLflow: {experiment_name}\")\n"," print(f\"{'='*60}\")\n","\n"," # 1. Filtrage des configurations à lancer\n"," if configs_to_test is None:\n"," remaining_configs = filter_configs_to_run(experimental_configurations, experiment_name)\n","\n"," if not remaining_configs:\n"," print(\"\\n🎉 TOUTES LES EXPÉRIENCES SONT DÉJÀ TERMINÉES !\")\n"," print(\"Rien à faire. Utilisez compare_experiments_mlflow() pour voir les résultats.\")\n"," return []\n"," if TEST_MODE:\n"," configs_to_test = remaining_configs[:TEST_MAX_CONFIGS] if TEST_MAX_CONFIGS else remaining_configs\n"," print(f\"\\n🧪 Mode TEST: limitation à {len(configs_to_test)} configurations\")\n"," else:\n"," configs_to_test = remaining_configs\n"," else:\n"," completed_runs = get_completed_experiments(experiment_name)\n"," filtered_configs = []\n"," for config_idx in configs_to_test:\n"," if config_idx >= len(experimental_configurations):\n"," print(f\"⚠️ Index {config_idx} invalide, ignoré\")\n"," continue\n"," config = experimental_configurations[config_idx]\n"," expected_run_name = generate_stable_run_name(config, TEST_MODE)\n"," if expected_run_name in completed_runs:\n"," print(f\"⏭️ [{config_idx}] {config['name']} déjà terminé, ignoré\")\n"," else:\n"," filtered_configs.append(config_idx)\n"," configs_to_test = filtered_configs\n","\n"," if not configs_to_test:\n"," print(\"\\n✅ Toutes les configurations demandées sont déjà terminées !\")\n"," return []\n","\n"," # 2. Boucle d'entraînement avec nettoyage SIMPLIFIÉ\n"," print(f\"\\n🎯 ENTRAÎNEMENT DE {len(configs_to_test)} CONFIGURATIONS:\")\n","\n"," results = []\n"," for i, config_idx in enumerate(configs_to_test):\n"," config = experimental_configurations[config_idx]\n"," stable_run_name = generate_stable_run_name(config, TEST_MODE)\n","\n"," print(f\"\\n{'-'*50}\")\n"," print(f\"🧪 Expérience {i+1}/{len(configs_to_test)}: {config['name']}\")\n"," print(f\"🏷️ Run name: {stable_run_name}\")\n"," print(f\"{'-'*50}\")\n","\n"," try:\n"," # ENTRAÎNEMENT (ici artifacts_dir existe et fonctionne normalement)\n"," result = train_segmentation_model_optimized(config, run_name=stable_run_name)\n"," results.append(result)\n"," print(f\"✅ [{config_idx}] {config['name']} terminé avec succès\")\n","\n"," remaining = len(configs_to_test) - (i + 1)\n"," if remaining > 0:\n"," print(f\"🔄 {remaining} configuration(s) restante(s)\")\n","\n"," except Exception as e:\n"," print(f\"❌ Erreur avec [{config_idx}] {config['name']}: {e}\")\n","\n"," # NETTOYAGE UNIQUE : après chaque modèle (réussi ou échoué)\n"," if i < len(configs_to_test) - 1: # Pas de nettoyage après le dernier modèle\n"," print(\"🧹 Nettoyage VRAM entre modèles...\")\n"," deep_clean_memory()\n","\n"," # 3. Résumé final\n"," print(f\"\\n{'='*60}\")\n"," print(f\"📊 RÉSUMÉ DE CETTE SESSION\")\n"," print(f\"{'='*60}\")\n","\n"," if results:\n"," print(f\"✅ {len(results)} nouvelles expériences réussies\")\n","\n"," print(f\"\\n{'Modèle':<25} {'Val Acc':<10} {'Test Acc':<10} {'Mean IoU':<10} {'Temps':<8} {'Run ID':<10}\")\n"," print(\"-\" * 80)\n","\n"," for result in results:\n"," config_name = result['config']['name']\n"," val_acc = result['best_val_accuracy']\n"," test_acc = result['test_metrics'].get('sparse_categorical_accuracy', 0)\n"," mean_iou = result['detailed_metrics']['mean_iou']\n"," time_min = result['training_time'] / 60\n"," run_id = result['run_id'][:8]\n"," print(f\"{config_name:<25} {val_acc:<10.4f} {test_acc:<10.4f} {mean_iou:<10.4f} {time_min:<8.1f} {run_id:<10}\")\n","\n"," best_result = max(results, key=lambda x: x['detailed_metrics']['mean_iou'])\n"," print(f\"\\n🏆 Meilleur de cette session: {best_result['config']['name']}\")\n"," print(f\" Val Acc: {best_result['best_val_accuracy']:.4f}\")\n"," print(f\" Test Acc: {best_result['test_metrics'].get('sparse_categorical_accuracy', 0):.4f}\")\n"," print(f\" Mean IoU: {best_result['detailed_metrics']['mean_iou']:.4f}\")\n"," print(f\" MLflow Run: {best_result['run_id']}\")\n","\n"," total_remaining = filter_configs_to_run(experimental_configurations, experiment_name)\n"," if len(total_remaining) > len(configs_to_test):\n"," remaining_count = len(total_remaining) - len(configs_to_test)\n"," print(f\"\\n🔄 Il reste encore {remaining_count} configuration(s) à entraîner\")\n"," print(\"💡 Relancez le script pour continuer automatiquement\")\n"," else:\n"," print(f\"\\n🎉 TOUTES LES CONFIGURATIONS SONT MAINTENANT TERMINÉES !\")\n"," print(\"💡 Utilisez compare_experiments_local() pour voir tous les résultats\")\n"," else:\n"," print(\"❌ Aucune nouvelle expérience réussie dans cette session\")\n","\n"," return results"]},{"cell_type":"code","execution_count":21,"metadata":{"id":"SdG8VxB7v8tF","executionInfo":{"status":"ok","timestamp":1754980948860,"user_tz":-120,"elapsed":9,"user":{"displayName":"Grégoire","userId":"01249298206117462766"}}},"outputs":[],"source":["# ====================================================================\n","# COMPARAISON DES EXPÉRIENCES LOCALES\n","# ====================================================================\n","\n","def compare_experiments_local(artifacts_base_dir=\"artifacts\"):\n"," \"\"\"\n"," Compare les résultats de plusieurs expériences via les artefacts locaux\n"," Structure : artifacts/run_{run_id}/\n"," \"\"\"\n"," if not os.path.exists(artifacts_base_dir):\n"," print(\"❌ Aucun répertoire d'artefacts trouvé\")\n"," return\n","\n"," results = []\n","\n"," print(f\"🔍 Recherche des expériences dans {artifacts_base_dir}/...\")\n","\n"," # Parcourir tous les répertoires run_*\n"," run_dirs = [d for d in os.listdir(artifacts_base_dir) if d.startswith('run_')]\n","\n"," if not run_dirs:\n"," print(\"❌ Aucune expérience trouvée\")\n"," return\n","\n"," print(f\"📊 {len(run_dirs)} expériences trouvées\")\n","\n"," for run_dir in run_dirs:\n"," run_path = os.path.join(artifacts_base_dir, run_dir)\n"," config_path = os.path.join(run_path, \"model_config.json\")\n"," history_path = os.path.join(run_path, \"training_history.json\")\n"," detailed_path = os.path.join(run_path, \"detailed_metrics.json\")\n","\n"," # Vérifier que les fichiers existent\n"," if os.path.exists(config_path) and os.path.exists(history_path):\n"," try:\n"," # Charger la configuration\n"," with open(config_path, 'r') as f:\n"," config = json.load(f)\n","\n"," # Charger l'historique d'entraînement\n"," with open(history_path, 'r') as f:\n"," history = json.load(f)\n","\n"," # Charger les métriques détaillées (si disponibles)\n"," detailed_metrics = None\n"," if os.path.exists(detailed_path):\n"," with open(detailed_path, 'r') as f:\n"," detailed_metrics = json.load(f)\n","\n"," # Extraire les métriques finales\n"," final_metrics = {\n"," 'run_id': run_dir.replace('run_', ''),\n"," 'experiment': config.get('name', 'Unknown'),\n"," 'architecture': config.get('architecture', 'Unknown'),\n"," 'encoder': config.get('encoder', 'None'),\n"," 'encoder_freeze': config.get('encoder_freeze', False),\n"," 'final_train_acc': history['sparse_categorical_accuracy'][-1],\n"," 'final_val_acc': history['val_sparse_categorical_accuracy'][-1],\n"," 'final_train_loss': history['loss'][-1],\n"," 'final_val_loss': history['val_loss'][-1],\n"," 'best_val_acc': max(history['val_sparse_categorical_accuracy']),\n"," 'epochs_trained': len(history['loss']),\n"," 'mean_iou': detailed_metrics['mean_iou'] if detailed_metrics else None,\n"," 'run_path': run_path\n"," }\n"," results.append(final_metrics)\n","\n"," except Exception as e:\n"," print(f\"⚠️ Erreur lors du chargement de {run_dir}: {e}\")\n"," continue\n","\n"," # Afficher les résultats\n"," if results:\n"," print(f\"\\n🏆 COMPARAISON DES {len(results)} EXPÉRIENCES\")\n"," print(\"=\" * 100)\n","\n"," # Trier par Mean IoU si disponible, sinon par best_val_acc\n"," if any(r['mean_iou'] is not None for r in results):\n"," results_sorted = sorted(results, key=lambda x: x['mean_iou'] or 0, reverse=True)\n"," print(\"📊 Classement par Mean IoU (meilleur premier)\")\n"," else:\n"," results_sorted = sorted(results, key=lambda x: x['best_val_acc'], reverse=True)\n"," print(\"📊 Classement par Validation Accuracy (meilleur premier)\")\n","\n"," print()\n","\n"," for i, result in enumerate(results_sorted, 1):\n"," print(f\"🥇 #{i} - {result['experiment']}\")\n"," print(f\" 🏗️ Architecture: {result['architecture']}\")\n","\n"," if result['encoder'] != 'None':\n"," freeze_status = \"🧊 Frozen\" if result['encoder_freeze'] else \"🔥 Trainable\"\n"," print(f\" 🧠 Encoder: {result['encoder']} ({freeze_status})\")\n","\n"," print(f\" 📊 Meilleure Val Acc: {result['best_val_acc']:.4f}\")\n"," print(f\" 📊 Val Acc finale: {result['final_val_acc']:.4f}\")\n","\n"," if result['mean_iou'] is not None:\n"," print(f\" 🎯 Mean IoU: {result['mean_iou']:.4f}\")\n","\n"," print(f\" ⏱️ Époques: {result['epochs_trained']}\")\n"," print(f\" 🆔 Run ID: {result['run_id'][:8]}...\")\n"," print(f\" 📁 Chemin: {result['run_path']}\")\n"," print()\n","\n"," # Statistiques globales\n"," print(\"📈 STATISTIQUES GLOBALES:\")\n"," print(\"-\" * 50)\n"," avg_val_acc = sum(r['best_val_acc'] for r in results) / len(results)\n"," print(f\"🎯 Val Acc moyenne: {avg_val_acc:.4f}\")\n","\n"," if any(r['mean_iou'] is not None for r in results):\n"," valid_ious = [r['mean_iou'] for r in results if r['mean_iou'] is not None]\n"," avg_iou = sum(valid_ious) / len(valid_ious)\n"," print(f\"🎯 Mean IoU moyen: {avg_iou:.4f}\")\n","\n"," avg_epochs = sum(r['epochs_trained'] for r in results) / len(results)\n"," print(f\"⏱️ Époques moyennes: {avg_epochs:.1f}\")\n","\n"," # Architecture summary\n"," architectures = {}\n"," for r in results:\n"," arch = r['architecture']\n"," if arch not in architectures:\n"," architectures[arch] = []\n"," architectures[arch].append(r['best_val_acc'])\n","\n"," print(f\"\\n🏗️ PERFORMANCE PAR ARCHITECTURE:\")\n"," print(\"-\" * 50)\n"," for arch, accs in architectures.items():\n"," avg_acc = sum(accs) / len(accs)\n"," print(f\"{arch}: {avg_acc:.4f} (moyenne sur {len(accs)} runs)\")\n","\n"," else:\n"," print(\"❌ Aucune expérience valide trouvée\")\n","\n"," return results"]},{"cell_type":"code","execution_count":22,"metadata":{"id":"DWf1lWhyv_OI","executionInfo":{"status":"ok","timestamp":1754980948936,"user_tz":-120,"elapsed":68,"user":{"displayName":"Grégoire","userId":"01249298206117462766"}}},"outputs":[],"source":["# ====================================================================\n","# COMPARAISON DES EXPÉRIENCES LOCALES\n","# ====================================================================\n","\n","def show_best_experiment_details(artifacts_base_dir=\"artifacts\"):\n"," \"\"\"\n"," Affiche les détails de la meilleure expérience\n"," \"\"\"\n"," results = compare_experiments_local(artifacts_base_dir)\n","\n"," if not results:\n"," return\n","\n"," # Meilleure expérience (Mean IoU si disponible, sinon Val Acc)\n"," if any(r['mean_iou'] is not None for r in results):\n"," best = max(results, key=lambda x: x['mean_iou'] or 0)\n"," metric_name = \"Mean IoU\"\n"," metric_value = best['mean_iou']\n"," else:\n"," best = max(results, key=lambda x: x['best_val_acc'])\n"," metric_name = \"Validation Accuracy\"\n"," metric_value = best['best_val_acc']\n","\n"," print(f\"\\n🏆 DÉTAILS DE LA MEILLEURE EXPÉRIENCE ({metric_name})\")\n"," print(\"=\" * 60)\n"," print(f\"🎯 Nom: {best['experiment']}\")\n"," print(f\"🏗️ Architecture: {best['architecture']}\")\n","\n"," if best['encoder'] != 'None':\n"," freeze_status = \"🧊 Frozen\" if best['encoder_freeze'] else \"🔥 Trainable\"\n"," print(f\"🧠 Encoder: {best['encoder']} ({freeze_status})\")\n","\n"," print(f\"📊 {metric_name}: {metric_value:.4f}\")\n"," print(f\"📊 Validation Accuracy: {best['best_val_acc']:.4f}\")\n"," print(f\"⏱️ Époques: {best['epochs_trained']}\")\n"," print(f\"🆔 Run ID: {best['run_id']}\")\n"," print(f\"📁 Chemin des artefacts: {best['run_path']}\")\n","\n"," # Vérifier la présence du modèle\n"," model_path = os.path.join(best['run_path'], 'best_model.keras')\n"," if os.path.exists(model_path):\n"," print(f\"✅ Modèle disponible: {model_path}\")\n"," else:\n"," print(f\"❌ Modèle non trouvé: {model_path}\")"]},{"cell_type":"code","execution_count":null,"metadata":{"executionInfo":{"elapsed":4943,"status":"aborted","timestamp":1754980887153,"user":{"displayName":"Grégoire","userId":"01249298206117462766"},"user_tz":-120},"id":"tlpWs7XqdCF5"},"outputs":[],"source":["# ====================================================================\n","# POINT D'ENTRÉE PRINCIPAL\n","# ====================================================================\n","\n","if __name__ == \"__main__\":\n"," print(\"🎯 SCRIPT D'ENTRAÎNEMENT OPTIMISÉ POUR SEGMENTATION\")\n"," print(f\"Mode: {'🧪 TEST' if TEST_MODE else '🚀 PRODUCTION'}\")\n"," print(\"🔄 SYSTÈME DE REPRISE AUTOMATIQUE ACTIVÉ\")\n","\n"," # Vérifications préliminaires\n"," try:\n"," print(f\"MLflow version: {mlflow.__version__}\")\n"," print(f\"Cityscapes dir: {cityscapes_dir}\")\n"," print(f\"Configurations disponibles: {len(experimental_configurations)}\")\n","\n"," assert callable(create_model_from_config), \"create_model_from_config non disponible\"\n"," assert callable(prepare_cityscapes_dataset), \"prepare_cityscapes_dataset non disponible\"\n","\n"," print(\"✅ Toutes les vérifications passées\")\n","\n"," # Lancer les expériences avec système de reprise\n"," results = run_experiments_optimized()\n","\n"," print(\"\\n🎉 Script terminé avec succès!\")\n","\n"," except Exception as e:\n"," print(f\"❌ Erreur fatale: {e}\")\n"," print(\"Vérifications:\")\n"," print(\"1. Variables cityscapes_dir et experimental_configurations définies ?\")\n"," print(\"2. Fonctions create_model_from_config, prepare_cityscapes_dataset disponibles ?\")\n"," print(\"3. MLflow configuré ?\")\n"," print(\"4. Données Cityscapes présentes ?\")"]},{"cell_type":"markdown","metadata":{"id":"6gJK8H-YRwRN"},"source":["## Analyse des expériences"]},{"cell_type":"markdown","metadata":{"id":"erK7RXCV7HOz"},"source":["### Comparaison des experiences"]},{"cell_type":"code","execution_count":29,"metadata":{"id":"LQqdr5KUTWZW","executionInfo":{"status":"ok","timestamp":1754980984980,"user_tz":-120,"elapsed":35,"user":{"displayName":"Grégoire","userId":"01249298206117462766"}}},"outputs":[],"source":["def get_completed_configs_info(experimental_configurations, experiment_name=\"OC Projet 8\"):\n"," \"\"\"\n"," Réutilise la logique de filter_configs_to_run mais retourne les configs terminées\n"," \"\"\"\n"," completed_runs = get_completed_experiments(experiment_name)\n","\n"," if not completed_runs:\n"," return []\n","\n"," # Même logique que filter_configs_to_run mais on garde les terminées\n"," completed_configs = []\n","\n"," for idx, config in enumerate(experimental_configurations):\n"," expected_run_name = generate_stable_run_name(config, TEST_MODE)\n","\n"," if expected_run_name in completed_runs: # Condition inversée\n"," completed_configs.append((idx, config, expected_run_name))\n","\n"," return completed_configs\n","\n","def compare_experiments_mlflow(experiment_name=\"OC Projet 8\", show_detailed_results=False, show_plots=True):\n"," \"\"\"\n"," Compare les expériences terminées - version ultra-simple\n","\n"," Args:\n"," experiment_name (str): Nom de l'expérience MLflow\n"," show_detailed_results (bool): Si True, affiche et retourne la liste complète des résultats\n"," show_plots (bool): Si True, génère les diagrammes en barres des métriques\n"," \"\"\"\n"," # Récupérer les configs terminées (réutilise la logique existante)\n"," completed_configs = get_completed_configs_info(experimental_configurations, experiment_name)\n","\n"," if not completed_configs:\n"," print(\"❌ Aucune expérience terminée trouvée\")\n"," return []\n","\n"," # Récupérer le mapping run_name -> run_id depuis MLflow\n"," experiment = mlflow.get_experiment_by_name(experiment_name)\n"," runs_df = mlflow.search_runs(\n"," experiment_ids=[experiment.experiment_id],\n"," filter_string=\"status = 'FINISHED'\",\n"," output_format=\"pandas\"\n"," )\n","\n"," run_name_to_id = {\n"," run.get('tags.mlflow.runName', ''): run['run_id']\n"," for _, run in runs_df.iterrows()\n"," }\n","\n"," print(f\"📊 Analyse de {len(completed_configs)} configurations terminées\")\n","\n"," results = []\n","\n"," for config_idx, config, run_name in completed_configs:\n"," if run_name not in run_name_to_id:\n"," continue\n","\n"," run_id = run_name_to_id[run_name]\n","\n"," try:\n"," # Récupérer les métriques directement depuis MLflow\n"," run = mlflow.get_run(run_id)\n"," metrics = run.data.metrics\n","\n"," # Récupérer l'historique d'entraînement pour certaines infos\n"," import tempfile\n"," with tempfile.TemporaryDirectory() as temp_dir:\n"," try:\n"," history_path = mlflow.artifacts.download_artifacts(\n"," run_id=run_id, artifact_path=\"training_history.json\", dst_path=temp_dir\n"," )\n"," with open(history_path, 'r') as f:\n"," history = json.load(f)\n"," best_val_acc = max(history['val_sparse_categorical_accuracy'])\n"," epochs_trained = len(history['loss'])\n"," except:\n"," # Fallback si l'artefact n'est pas disponible\n"," best_val_acc = metrics.get('val_sparse_categorical_accuracy', 'N/A')\n"," epochs_trained = 'N/A'\n","\n"," # Construire le résultat avec les métriques de test depuis MLflow\n"," result = {\n"," 'config_idx': config_idx,\n"," 'experiment': config['name'],\n"," 'architecture': config.get('architecture', 'Unknown'),\n"," 'encoder': config.get('encoder', 'None'),\n"," 'best_val_acc': best_val_acc,\n"," 'test_mean_iou': metrics.get('test_mean_iou', 'N/A'),\n"," 'test_accuracy': metrics.get('test_sparse_categorical_accuracy', 'N/A'),\n"," 'epochs_trained': epochs_trained,\n"," 'run_name': run_name,\n"," 'run_id': run_id,\n"," 'full_config': config # Toujours inclure la config complète\n"," }\n","\n"," results.append(result)\n","\n"," except Exception as e:\n"," print(f\"⚠️ Erreur pour [{config_idx}] {config['name']}: {e}\")\n"," continue\n","\n"," # Affichage\n"," if results:\n"," # Trier par test_mean_iou si disponible, sinon par best_val_acc\n"," def sort_key(x):\n"," if x['test_mean_iou'] != 'N/A':\n"," return x['test_mean_iou']\n"," return x['best_val_acc']\n","\n"," results_sorted = sorted(results, key=sort_key, reverse=True)\n","\n"," print(f\"\\n🏆 TOP {len(results)} EXPÉRIENCES TERMINÉES\")\n"," print(\"=\" * 100)\n","\n"," for i, result in enumerate(results_sorted, 1):\n"," print(f\"🥇 #{i} - [{result['config_idx']}] {result['experiment']}\")\n"," print(f\" 🏗️ {result['architecture']} + {result['encoder']}\")\n","\n"," # Afficher les métriques de test\n"," test_iou = result['test_mean_iou']\n"," test_acc = result['test_accuracy']\n","\n"," if test_iou != 'N/A':\n"," print(f\" 🎯 Test Mean IoU: {test_iou:.4f}\", end=\"\")\n"," else:\n"," print(f\" 🎯 Test Mean IoU: N/A\", end=\"\")\n","\n"," if test_acc != 'N/A':\n"," print(f\" | Test Acc: {test_acc:.4f}\")\n"," else:\n"," print(f\" | Test Acc: N/A\")\n","\n"," print(f\" 🆔 {result['run_id'][:8]}...\")\n","\n"," # Afficher la configuration complète si demandé\n"," if show_detailed_results:\n"," print(f\" 📋 Configuration complète:\")\n"," import pprint\n"," pprint.pprint(result['full_config'], indent=6)\n","\n"," print()\n","\n"," # Génération des diagrammes si demandé\n"," if show_plots and results:\n"," generate_comparison_plots(results_sorted)\n","\n"," # Retourner les résultats seulement si explicitement demandé\n"," if show_detailed_results:\n"," return results\n"," else:\n"," print(f\"\\n✅ Analyse terminée. Utilisez show_detailed_results=True pour obtenir les données détaillées.\")\n"," return None\n","\n","\n","def generate_comparison_plots(results):\n"," \"\"\"\n"," Génère des diagrammes en barres pour comparer les métriques de test\n","\n"," Args:\n"," results (list): Liste des résultats triés\n"," \"\"\"\n"," import matplotlib.pyplot as plt\n"," import numpy as np\n","\n"," # Filtrer les résultats qui ont des métriques de test valides\n"," valid_results_iou = [r for r in results if r['test_mean_iou'] != 'N/A']\n"," valid_results_acc = [r for r in results if r['test_accuracy'] != 'N/A']\n","\n"," # Configuration des graphiques\n"," plt.style.use('seaborn-v0_8' if 'seaborn-v0_8' in plt.style.available else 'default')\n","\n"," # Créer une figure avec 2 sous-graphiques\n"," fig, (ax1, ax2) = plt.subplots(1, 2, figsize=(20, 8))\n","\n"," # Graphique 1: Test Mean IoU\n"," if valid_results_iou:\n"," experiment_names = [r['experiment'] for r in valid_results_iou]\n"," iou_values = [r['test_mean_iou'] for r in valid_results_iou]\n","\n"," # Créer des labels courts pour les noms d'expériences\n"," short_labels = []\n"," for name in experiment_names:\n"," # Raccourcir les noms longs\n"," if len(name) > 25:\n"," parts = name.split('_')\n"," short_name = f\"{parts[0]}_{parts[1] if len(parts) > 1 else ''}...\"\n"," short_labels.append(short_name)\n"," else:\n"," short_labels.append(name)\n","\n"," colors = plt.cm.viridis(np.linspace(0, 1, len(iou_values)))\n"," bars1 = ax1.bar(range(len(iou_values)), iou_values, color=colors, alpha=0.8)\n","\n"," ax1.set_title('Test Mean IoU par Expérience', fontsize=14, fontweight='bold')\n"," ax1.set_xlabel('Expériences', fontsize=12)\n"," ax1.set_ylabel('Mean IoU', fontsize=12)\n"," ax1.set_xticks(range(len(short_labels)))\n"," ax1.set_xticklabels(short_labels, rotation=45, ha='right')\n"," ax1.grid(True, alpha=0.3)\n","\n"," # Ajouter les valeurs sur les barres\n"," for bar, value in zip(bars1, iou_values):\n"," height = bar.get_height()\n"," ax1.text(bar.get_x() + bar.get_width()/2., height + 0.001,\n"," f'{value:.4f}', ha='center', va='bottom', fontsize=10)\n"," else:\n"," ax1.text(0.5, 0.5, 'Aucune donnée Mean IoU disponible',\n"," ha='center', va='center', transform=ax1.transAxes, fontsize=14)\n"," ax1.set_title('Test Mean IoU par Expérience', fontsize=14, fontweight='bold')\n","\n"," # Graphique 2: Test Accuracy\n"," if valid_results_acc:\n"," experiment_names = [r['experiment'] for r in valid_results_acc]\n"," acc_values = [r['test_accuracy'] for r in valid_results_acc]\n","\n"," # Créer des labels courts pour les noms d'expériences\n"," short_labels = []\n"," for name in experiment_names:\n"," if len(name) > 25:\n"," parts = name.split('_')\n"," short_name = f\"{parts[0]}_{parts[1] if len(parts) > 1 else ''}...\"\n"," short_labels.append(short_name)\n"," else:\n"," short_labels.append(name)\n","\n"," colors = plt.cm.plasma(np.linspace(0, 1, len(acc_values)))\n"," bars2 = ax2.bar(range(len(acc_values)), acc_values, color=colors, alpha=0.8)\n","\n"," ax2.set_title('Test Accuracy par Expérience', fontsize=14, fontweight='bold')\n"," ax2.set_xlabel('Expériences', fontsize=12)\n"," ax2.set_ylabel('Accuracy', fontsize=12)\n"," ax2.set_xticks(range(len(short_labels)))\n"," ax2.set_xticklabels(short_labels, rotation=45, ha='right')\n"," ax2.grid(True, alpha=0.3)\n","\n"," # Ajouter les valeurs sur les barres\n"," for bar, value in zip(bars2, acc_values):\n"," height = bar.get_height()\n"," ax2.text(bar.get_x() + bar.get_width()/2., height + 0.001,\n"," f'{value:.4f}', ha='center', va='bottom', fontsize=10)\n"," else:\n"," ax2.text(0.5, 0.5, 'Aucune donnée Accuracy disponible',\n"," ha='center', va='center', transform=ax2.transAxes, fontsize=14)\n"," ax2.set_title('Test Accuracy par Expérience', fontsize=14, fontweight='bold')\n","\n"," plt.tight_layout()\n"," plt.show()\n","\n"," print(f\"📊 Diagrammes générés pour {len(valid_results_iou)} expériences (IoU) et {len(valid_results_acc)} expériences (Accuracy)\")"]},{"cell_type":"code","execution_count":30,"metadata":{"colab":{"base_uri":"https://localhost:8080/","height":1000,"referenced_widgets":["1091b77c2b2c415ca0d627933b64969c","43eff508df454d69a87ebab244700e9e","c12fb4498f294c35bd4e1ac67823a534","5c7fa581589741dda94baccef1f9564d","f0a3e51745994b50ba0e783265c32137","86680a798bba4ade89097aff42d7570b","aec87ffacce74aebb2067b45382c6d1c","7e2c871cfb284873b63de0e836f81b8b","93c0a6cd576c462caf536af285270ffc","57fad05e86024b8087b1b34b69aa5a91","ea302d170c65484c98efabf85f07af46","3f4fde907ea74cd387b8586573cdd9db","09a106f167184b478af1366bc1185a7a","2e73c418443c4dbf9cdd81f34d5b25d8","78b28795a0904ef7b426bfe641200eb1","f0dbde1d6a79441f9e2d3aaa95f3f33b","1fd4732a61fb48a1ba15754105557f90","a760121efac043d79a376c32bd930c11","4599d7c666b846b5b59ea43aac1aaa66","93e95dfcb3e54c01af548b40e20a5c2b","ade00d01b1c14087ae014c1f6a528789","ed9faf1158b74459b91fd1279b4f006c","1395bc9b8fff44fa8a0491f2f71b02e2","171cf780a6b543e5bcb2dbb12c47bb7d","5ebab0c35f5f41ab8c9e87216e8a2720","e94cb923a41b436f8ddc027f2f0024f5","615e04cf1c0941e8856218aa913fd8b9","0ea2448f650c41f999072555710933d7","5cee1b01776f456c8236f7f9bf9773da","897fc5fa5b494dbdbcdbaff6df34125e","b4becb9a0f584b389fd9e6c30710e2f5","362d49dbf9c84c5fa94f646925d7b091","6560c3d99f434eff99489626006fce6e","2e51f1811257441c9a6aae73e4d7dd7a","5385040a9cb6479fba024d3c44718f0f","ba1c2764e94e41fc84f209316cdcd8e7","0afb88a8f76743dbac73b6120a652fb3","eeb506b7031e4faa99c13156574ce849","b03b36a3cda343eab4302f9800666a09","2e48d976d2994f02a9c6c354ded6d93d","7c5d60d4391d4f69bdcc4606a0040418","d62525a9065a4d8ab1e7f13341fd9f31","08127b71707b46c59bebdf30ed881aff","9cb723f820574e82a7e7043a0b667a2e","6313cdcb83cf47bbb66b1d036636e80c","98756ebad2814918a0ed2364ef43a3c3","a3122178c05e4fe6b9050006206b0757","f6dae4271504411f82af9b2c9ba2a9a6","60337d9c340541bca4b5364e2cd3ad70","cbd4745db3964c3597b3b5cef3a7a831","526c741e78ef475bab96a2be655a4218","9970a50198804a1ca1497ee01b4dc483","65cec6c7efcc4cbc961620366b397348","e3633532b4bc40789552749f3dc191b5","b8d30a1e85c94a119d51602971c50edf","cbe21982571a4d25945f98ae4bce71cc","b386e9ac989e42f189f22356e3f36490","92cbe305e54b4961820a692c5cc03ae4","2c4bea6986f3493e82006a492c5cf7f3","c7421ae292a24dc7a96810e2a5b33fd4","c3d523bab420406fabcf111322c918e7","7c5659046dda43039d28bd62b126316c","7ce165d634184ed1a425bcdbc5fda1b7","ae5e3ed525bf4615a3248263d14ee9bb","0f605c65e65e4adcae52c58ffbe37c3c","cf2538490fa749008e0d957d20352d3d","e99370e5788c46d0b39795d99dee61fb","4374864e0b5d42df8e0b43723161c3c0","1de35ebd7d65431caa54aef21449abd7","02cdca3d4e924090aeba15d1798ce7bd","94afa7b84173410cb6f958c9ccd0422c","58d833f900f0430abcdb0c78356e3214","9fed4393c78e4e0fa5e482e3f4b69f46","14e3aa699f304caca909af1425a52bc6","ce676fee299b4b30ae9fd35fd81363f1","d3822fb07cb54622924f59f9f523550e","df856305989145368385950261150eed","fc2df5b47eb24f4087a4503cffb77246","eb0fa1515aec4d8c8a8ae86949b13d6e","2e31213718b44a3f9f9a31ce9bca6ff8","83f75e069ffe42e9853dd4abdec96687","fc3e7c7b77ab4179b32a9530b8520167","fdad39fd29a945559948f5a2e4df4f51","89c16600214447de83e453fe1f704193","c74c18bec4bc46828c642f822be77257","21f23e2a16ac4c3a822fb654f010ead9","7e3264726bef461686d0d1af54afd20d","1c4f5dc9514a4c2caf633b19c7d6f498","e558c3602e324ec08f4f2515e18e72c3","1614bb32193946599749936bde46a78b","147a96c1c54a47cbb306bdad580eae63","636fd1f3496c4359ae248e9ffa108044","21aa609507034287a2f67ddd67c88f9b","31666151387a4e13b24279c4319c42ce","6864f54445a04f0294e7d5a6bc7e0b9c","3c73a65abb8f4785a2225273c4cb6700","c1c34817a6764921ada246a927bcf605","22668934797b43298e4b1dba5534002f","62f3d49ddcbf4e04821039f06f9f44a5","d2b85c571b654d5c830954973fbf969b","7c277e3b8f304008b584db188a55565e","bdc784c43f534f608e75a51e450a2469","740df50f8c7045f19599645a59ebf050","7a991ad79c5a4c8fa0ed4ad9685ca49e","7c81408de8f34579bf4050e9d9c605d9","77620217a9af40c392c9df32d324cb9d","56fffbaf1902493b8403a5c6805b4081","2d13cefdfa0a4284b35cac3f6e1ed5d5","2a2b571a62214acba786b8b983221b87","c85a81534e704418bf7bb9a78e03fbc8","9ea49468c9c44d0db2b64def1a435387","cbdc4d6c8ba2498294615949689750c9","569404e538264e5cb42a23990c01af95","eee1b08234fb4050aa0b484f985bf610","faad9784703b40c182326bbcbacf7dca","e7aac0ec04e4468aaa160560682349ed","688785d14b4a47b3a97125199f2eb4b3","1c632146bae8449d811dd1de87f5f003","16523bcd0c7e4c789ad76e8e35daa401","4783d7d77ab44eeb994ca3a377155881","7e59846df494492a8a0f271a8d7bbec2","a6a9e50c210f4daeb9b6fb748432a2c1","608ed2ec0db84db390718595ca5874e0","51c6225409054c4c92b2ede2786c0fdd","61515b9d53aa496a92a35317b0217399","a7d8e443753a4e73aa979a3d947ecf58","69f7e0b1cdf04253a9084548f8a68313","edccc238492344a79410abf0c9262f42","5eccd0148c884158998ea322483314b9","dce2ed3f516346e7aa052adb7a2e4a8b","72ee36ebadcb4352bb7d6bea912663b3","bdfb780bb268497eb516cfc0cff5a1c7","4acaceb9c30941c9b7aa7374915af84c","f83673567ce445d68276fe036576f4d4","10a5557d53644b53988e93b8fa18bcef","48eba375b08e4355bd976cf8613fd1ed","e94f834bd244488e93ac95484b3bce89","3e8d88e4bdb843e59cf4b6e5d1900f72","c28d360f92624872aed1cef476834911","f1d6e5f5708b4b87ad8fc898f1f53dc3","618056d1b8034c2ab934a70f1477456e","759e24f886654d15afc5abbe8ff94222","3078b27db8714476ba11dc6d16918e67","bb6ae55e417c49e282067c699c17bfd8","bd04d00d2bd54da5981fd0b7c0c70289","2d15be12b43f4b78983f8fbd794412ca","f0b65f925a3b4faf9140bec936756d67","606aabe3bcaa46c3a94eaa3ae6fe4b25","782e32e38739446db73c9dbe39537de6","4d6884c1140b4a619cd24bdea8f947cf","2b8c4d2e9e224488aa611e32f829514b","96d9cae1f5644ccd88f32954e608634e","aa26f841eb1f4ed69062212afbc9a8e1","c445b8acbbc04933b55141d89490f2a3","f31964f7836344d88cd57b022ac70153","7d40c393476c4723a3021e008c78c266","bd591292aedf44d7bf7a1d86f3d738ca","27b3823545174d948fd9a160a8c077fa","620d915e5a4e4e258b54d5fa5f037791","3444956450d54f069e990a9a7ad492f0","7ca303aa42d4440f8499cc6b989940eb","42b3280f0b2c41e6bc0e20fab3ab3ad6","ffc5c47b83584884b37261489f7c0e5c","40f905a5742c48aba5d16b6e821fdcd5","f7cd64a5f6454cf993311adc7adf7e12","83425813fffd455a9611208bb3a60b06","4ecd130fcaa946f8b48165e496877b64","09748ce429eb458db4ad8700eb3c4f08","d0a2e8ce63e840139268a4ac520011c7","c9eac444afdc40f4ad25a4490b104c4f","3f07e49458ca4b039cbee102d1d490a3","c4d14d31830340c591255dea867fd0a0","e62f23222a974069b90b2d6bb576cfc0","2ff5895885ae479286f7f466191afc5c","fa203c08506c4b578fa77dce29e288c1","03a0a5fcd495465a8f7248e8f7abeca8","e70c0e6aa2fc43dba65d61d930c213eb","5add62ad858841ed925d812bc9441559","34ff4dcc63b1413b8bf3dd109eee1645","feb0ca5002fc4d7e9fa4e40f89523f7a","2b943d600c5046098227fa0255c1499c","3a06dc4c3f6242598721d483f298b939","84da2d7f03d74b57acd04f5218f76a9e","6c334e6792b048a8b12d72cd20c3db03","ef468e70a3a7419eb63f99d40e891846","ad33e4dba9144d28a3308fb903ecb639","8f525f792bda443abcb2ff887b9aa689"]},"executionInfo":{"elapsed":23506,"status":"ok","timestamp":1754981009244,"user":{"displayName":"Grégoire","userId":"01249298206117462766"},"user_tz":-120},"id":"fcl3B8x7PPrc","outputId":"cf628a0f-e4a2-4fc9-daa8-4f20a5da01ab"},"outputs":[{"output_type":"stream","name":"stdout","text":["🔍 Vérification des expériences déjà terminées dans 'OC Projet 8'...\n","✅ 39 expériences terminées trouvées:\n"," - FPN_efficientnetb0_finetune_augTrue_prod\n"," - FPN_efficientnetb0_finetune_augTrue_test\n"," - FPN_efficientnetb0_frozen_augFalse_prod\n"," - FPN_efficientnetb0_frozen_augFalse_test\n"," - FPN_efficientnetb0_frozen_augTrue_prod\n"," - FPN_efficientnetb0_frozen_augTrue_test\n"," - FPN_mobilenetv2_frozen_augFalse_prod\n"," - FPN_mobilenetv2_frozen_augFalse_test\n"," - FPN_mobilenetv2_frozen_augTrue_prod\n"," - FPN_mobilenetv2_frozen_augTrue_test\n"," - FPN_resnet34_finetune_augTrue_prod\n"," - FPN_resnet34_finetune_augTrue_test\n"," - FPN_resnet34_frozen_augFalse_prod\n"," - FPN_resnet34_frozen_augFalse_test\n"," - FPN_resnet34_frozen_augTrue_prod\n"," - FPN_resnet34_frozen_augTrue_test\n"," - FPN_vanilla_augFalse_prod\n"," - FPN_vanilla_augFalse_test\n"," - FPN_vanilla_augTrue_prod\n"," - FPN_vanilla_augTrue_test\n"," - U-Net_efficientnetb0_frozen_augFalse_prod\n"," - U-Net_efficientnetb0_frozen_augFalse_test\n"," - U-Net_efficientnetb0_frozen_augTrue_prod\n"," - U-Net_efficientnetb0_frozen_augTrue_test\n"," - U-Net_mobilenetv2_finetune_augTrue_prod\n"," - U-Net_mobilenetv2_finetune_augTrue_test\n"," - U-Net_mobilenetv2_frozen_augFalse_prod\n"," - U-Net_mobilenetv2_frozen_augFalse_test\n"," - U-Net_mobilenetv2_frozen_augTrue_prod\n"," - U-Net_mobilenetv2_frozen_augTrue_test\n"," - U-Net_vanilla_augFalse_prod\n"," - U-Net_vanilla_augFalse_prod_20250730_153443\n"," - U-Net_vanilla_augFalse_test\n"," - U-Net_vanilla_augTrue_prod\n"," - U-Net_vanilla_augTrue_prod_20250730_170602\n"," - U-Net_vanilla_augTrue_test\n"," - exp_001_baseline\n"," - exp_002_no_augmentation\n"," - exp_002_unfrozen\n","📊 Analyse de 17 configurations terminées\n"]},{"output_type":"display_data","data":{"text/plain":["Downloading artifacts: 0%| | 0/1 [00:00"],"image/png":"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\n"},"metadata":{}},{"output_type":"stream","name":"stdout","text":["📊 Diagrammes générés pour 17 expériences (IoU) et 17 expériences (Accuracy)\n","\n","✅ Analyse terminée. Utilisez show_detailed_results=True pour obtenir les données détaillées.\n"]}],"source":["compare_experiments_mlflow(experiment_name=mlflow_experiemnt)"]},{"cell_type":"markdown","metadata":{"id":"gqrw8X8wR5Q6"},"source":["### Analyse meilleur modèle"]},{"cell_type":"markdown","source":["Affichage des courbes d'apprentissage"],"metadata":{"id":"j9kVdhxVBA4S"}},{"cell_type":"code","source":["def display_training_curves_from_best_run(experiment_name=\"OC Projet 8\"):\n"," \"\"\"\n"," Trouve le meilleur modèle (selon test_mean_iou) et affiche son graphique de courbes d'entraînement.\n"," \"\"\"\n"," print(f\"\\n{'='*60}\")\n"," print(f\"📈 Affichage des courbes d'entraînement du meilleur modèle\")\n"," print(f\"🔬 Expérience MLflow: {experiment_name}\")\n"," print(f\"{'='*60}\")\n","\n"," client = mlflow.tracking.MlflowClient()\n","\n"," try:\n"," # Récupérer l'expérience par nom\n"," experiment = client.get_experiment_by_name(experiment_name)\n"," if experiment is None:\n"," print(f\"❌ Expérience '{experiment_name}' non trouvée.\")\n"," return\n","\n"," # Rechercher les runs terminés avec la métrique test_mean_iou\n"," runs_df = mlflow.search_runs(\n"," experiment_ids=[experiment.experiment_id],\n"," filter_string=\"status = 'FINISHED' and metrics.test_mean_iou >= 0\",\n"," order_by=[\"metrics.test_mean_iou DESC\"], # Trier par Mean IoU descendant\n"," max_results=1 # Prendre seulement le meilleur\n"," )\n","\n"," if runs_df.empty:\n"," print(\"❌ Aucun run terminé avec la métrique test_mean_iou trouvé.\")\n"," return\n","\n"," # Le meilleur run est le premier dans la liste triée\n"," best_run = runs_df.iloc[0]\n"," best_run_id = best_run['run_id']\n"," best_run_name = best_run.get('tags.mlflow.runName', 'Unknown')\n"," best_iou = best_run['metrics.test_mean_iou']\n","\n"," print(f\"🏆 Meilleur run trouvé:\")\n"," print(f\" • Run Name: {best_run_name}\")\n"," print(f\" • Run ID: {best_run_id}\")\n"," print(f\" • Test Mean IoU: {best_iou:.4f}\")\n","\n"," # Télécharger l'artefact 'training_curves.png'\n"," artifact_path = \"training_curves.png\"\n"," print(f\"\\n💾 Téléchargement de l'artefact '{artifact_path}'...\")\n","\n"," with tempfile.TemporaryDirectory() as tmpdir:\n"," local_path = mlflow.artifacts.download_artifacts(\n"," run_id=best_run_id,\n"," artifact_path=artifact_path,\n"," dst_path=tmpdir\n"," )\n"," print(f\"✅ Artefact téléchargé vers: {local_path}\")\n","\n"," # Afficher l'image\n"," print(\"\\n🖼️ Affichage de l'image...\")\n"," img = PILImage.open(local_path)\n"," plt.figure(figsize=(12, 8))\n"," plt.imshow(img)\n"," plt.title(f\"Courbes d'entraînement - {best_run_name}\", fontsize=14)\n"," plt.axis('off') # Désactiver les axes\n"," plt.show()\n","\n"," except Exception as e:\n"," print(f\"❌ Erreur lors de l l'affichage des courbes d'entraînement: {e}\")\n"," print(\"Vérifiez que l'artefact 'training_curves.png' existe pour le meilleur run.\")"],"metadata":{"id":"cQMW_Ri1dTE0","executionInfo":{"status":"ok","timestamp":1754981009265,"user_tz":-120,"elapsed":17,"user":{"displayName":"Grégoire","userId":"01249298206117462766"}}},"execution_count":31,"outputs":[]},{"cell_type":"code","source":["# Exécuter la fonction\n","display_training_curves_from_best_run(experiment_name=mlflow_experiemnt)"],"metadata":{"colab":{"base_uri":"https://localhost:8080/","height":948,"referenced_widgets":["83da010cc5a246eba6831ce173b91fa7","dcb54c2dfff645d9bec25e2b0c0ac850","c0769fc3f1834f8591441f82e389709b","ce5b51311b7b4c0c8f60fc222b0ece05","87b61be4b48646479a2fc73830237beb","4ae37cd8e3f84f99b61b3f8c0f4cd0d2","1bc2de107b774829ba2eea49c97835b0","9dff1f6de727469889f08dece2fad13c","7205672c3a644f4f9e513e05683f72c7","383eb8f5f0ed48b395503bd882b50021","fe138dece4e142c891526c1375667737"]},"id":"K7FA2AlTg0X-","executionInfo":{"status":"ok","timestamp":1754981013482,"user_tz":-120,"elapsed":4197,"user":{"displayName":"Grégoire","userId":"01249298206117462766"}},"outputId":"62cc0bfa-52d4-476c-bfe7-edc9e8f4b7bb"},"execution_count":32,"outputs":[{"output_type":"stream","name":"stdout","text":["\n","============================================================\n","📈 Affichage des courbes d'entraînement du meilleur modèle\n","🔬 Expérience MLflow: OC Projet 8\n","============================================================\n","🏆 Meilleur run trouvé:\n"," • Run Name: FPN_efficientnetb0_finetune_augTrue_prod\n"," • Run ID: b6dd1ecdeb3f4b5c899b332f3c948cd7\n"," • Test Mean IoU: 0.7457\n","\n","💾 Téléchargement de l'artefact 'training_curves.png'...\n"]},{"output_type":"display_data","data":{"text/plain":["Downloading artifacts: 0%| | 0/1 [00:00"],"image/png":"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\n"},"metadata":{}}]},{"cell_type":"markdown","source":["Affichage de la matrice de confusion en pourcentages"],"metadata":{"id":"GFqCXkZEBMp2"}},{"cell_type":"code","execution_count":33,"metadata":{"id":"YNQDzY0BH-X0","executionInfo":{"status":"ok","timestamp":1754981033304,"user_tz":-120,"elapsed":88,"user":{"displayName":"Grégoire","userId":"01249298206117462766"}}},"outputs":[],"source":["def display_best_confusion_matrix(experiment_name=\"OC Projet 8\"):\n"," \"\"\"\n"," Trouve le modèle avec la meilleure valeur test_mean_iou et affiche sa matrice de confusion\n","\n"," Args:\n"," experiment_name (str): Nom de l'expérience MLflow\n"," \"\"\"\n"," import matplotlib.pyplot as plt\n"," import seaborn as sns\n"," import numpy as np\n"," import json\n"," import tempfile\n","\n"," # Récupérer les résultats des expériences\n"," completed_configs = get_completed_configs_info(experimental_configurations, experiment_name)\n","\n"," if not completed_configs:\n"," print(\"❌ Aucune expérience terminée trouvée\")\n"," return None\n","\n"," # Récupérer le mapping run_name -> run_id depuis MLflow\n"," experiment = mlflow.get_experiment_by_name(experiment_name)\n"," runs_df = mlflow.search_runs(\n"," experiment_ids=[experiment.experiment_id],\n"," filter_string=\"status = 'FINISHED'\",\n"," output_format=\"pandas\"\n"," )\n","\n"," run_name_to_id = {\n"," run.get('tags.mlflow.runName', ''): run['run_id']\n"," for _, run in runs_df.iterrows()\n"," }\n","\n"," print(f\"🔍 Recherche du meilleur modèle parmi {len(completed_configs)} configurations...\")\n","\n"," best_result = None\n"," best_iou = -1\n","\n"," # Trouver le meilleur modèle\n"," for config_idx, config, run_name in completed_configs:\n"," if run_name not in run_name_to_id:\n"," continue\n","\n"," run_id = run_name_to_id[run_name]\n","\n"," try:\n"," # Récupérer les métriques\n"," run = mlflow.get_run(run_id)\n"," metrics = run.data.metrics\n","\n"," test_mean_iou = metrics.get('test_mean_iou', None)\n","\n"," if test_mean_iou is not None and test_mean_iou > best_iou:\n"," best_iou = test_mean_iou\n"," best_result = {\n"," 'config_idx': config_idx,\n"," 'experiment': config['name'],\n"," 'architecture': config.get('architecture', 'Unknown'),\n"," 'encoder': config.get('encoder', 'None'),\n"," 'test_mean_iou': test_mean_iou,\n"," 'test_accuracy': metrics.get('test_sparse_categorical_accuracy', 'N/A'),\n"," 'run_name': run_name,\n"," 'run_id': run_id\n"," }\n","\n"," except Exception as e:\n"," print(f\"⚠️ Erreur lors de l'analyse de {run_name}: {e}\")\n"," continue\n","\n"," if best_result is None:\n"," print(\"❌ Aucun modèle avec test_mean_iou trouvé\")\n"," return None\n","\n"," print(f\"🏆 MEILLEUR MODÈLE TROUVÉ:\")\n"," print(f\" 📊 [{best_result['config_idx']}] {best_result['experiment']}\")\n"," print(f\" 🏗️ {best_result['architecture']} + {best_result['encoder']}\")\n"," print(f\" 🎯 Test Mean IoU: {best_result['test_mean_iou']:.4f}\")\n"," print(f\" 📈 Test Accuracy: {best_result['test_accuracy']:.4f}\")\n"," print(f\" 🆔 Run ID: {best_result['run_id']}\")\n"," print()\n","\n"," # Télécharger et charger la matrice de confusion\n"," try:\n"," with tempfile.TemporaryDirectory() as temp_dir:\n"," confusion_matrix_path = mlflow.artifacts.download_artifacts(\n"," run_id=best_result['run_id'],\n"," artifact_path=\"confusion_matrix.json\",\n"," dst_path=temp_dir\n"," )\n","\n"," with open(confusion_matrix_path, 'r') as f:\n"," confusion_data = json.load(f)\n","\n"," matrix = np.array(confusion_data['matrix'])\n"," class_names = confusion_data['class_names']\n"," accuracy = confusion_data.get('accuracy', 'N/A')\n"," total_samples = confusion_data.get('total_samples', 'N/A')\n","\n"," print(f\"📋 Informations sur la matrice de confusion:\")\n"," print(f\" 🔢 Nombre de classes: {len(class_names)}\")\n"," print(f\" 📊 Échantillons totaux: {total_samples:,}\" if total_samples != 'N/A' else \" 📊 Échantillons totaux: N/A\")\n"," print(f\" ✅ Précision globale: {accuracy:.4f}\" if accuracy != 'N/A' else \" ✅ Précision globale: N/A\")\n"," print()\n","\n"," # Calculer les métriques par classe\n"," print(\"📈 Métriques par classe:\")\n"," print(\"-\" * 50)\n","\n"," for i, class_name in enumerate(class_names):\n"," # Précision: TP / (TP + FP)\n"," tp = matrix[i, i]\n"," fp = np.sum(matrix[:, i]) - tp\n"," fn = np.sum(matrix[i, :]) - tp\n","\n"," precision = tp / (tp + fp) if (tp + fp) > 0 else 0\n"," recall = tp / (tp + fn) if (tp + fn) > 0 else 0\n"," f1 = 2 * (precision * recall) / (precision + recall) if (precision + recall) > 0 else 0\n","\n"," print(f\" {class_name:12s} - Précision: {precision:.4f} | Rappel: {recall:.4f} | F1: {f1:.4f}\")\n","\n"," print()\n","\n"," # Créer la visualisation\n"," plt.figure(figsize=(12, 10))\n","\n"," # Normaliser la matrice pour l'affichage (en pourcentages)\n"," matrix_normalized = matrix.astype('float') / matrix.sum(axis=1)[:, np.newaxis] * 100\n","\n"," # Créer le heatmap\n"," sns.heatmap(matrix_normalized,\n"," annot=True,\n"," fmt='.1f',\n"," cmap='Blues',\n"," xticklabels=class_names,\n"," yticklabels=class_names,\n"," cbar_kws={'label': 'Pourcentage (%)'})\n","\n"," plt.title(f'Matrice de Confusion - {best_result[\"experiment\"]}\\n'\n"," f'Mean IoU: {best_result[\"test_mean_iou\"]:.4f} | Accuracy: {accuracy:.4f}',\n"," fontsize=14, fontweight='bold', pad=20)\n"," plt.xlabel('Classes Prédites', fontsize=12)\n"," plt.ylabel('Classes Réelles', fontsize=12)\n"," plt.xticks(rotation=45, ha='right')\n"," plt.yticks(rotation=0)\n","\n"," plt.tight_layout()\n"," plt.show()\n","\n"," # Optionnel: afficher aussi la matrice brute\n"," print(\"🔢 Matrice de confusion brute (premiers/derniers éléments):\")\n"," print(\"=\" * 60)\n"," print(\"Classes:\", class_names)\n"," print(\"Matrice (format: lignes=réel, colonnes=prédit):\")\n"," print(matrix)\n","\n"," return {\n"," 'confusion_matrix': matrix,\n"," 'class_names': class_names,\n"," 'best_model_info': best_result,\n"," 'accuracy': accuracy,\n"," 'total_samples': total_samples\n"," }\n","\n"," except Exception as e:\n"," print(f\"❌ Erreur lors du téléchargement de la matrice de confusion: {e}\")\n"," print(f\" Run ID: {best_result['run_id']}\")\n"," print(\" Vérifiez que l'artefact 'confusion_matrix.json' existe pour ce run.\")\n"," return None"]},{"cell_type":"code","execution_count":34,"metadata":{"colab":{"base_uri":"https://localhost:8080/","height":1000,"referenced_widgets":["ebe07ac726f2439ea7601b61c7d8ad08","6677b53645214d5d93d74b82956c55f6","dbc64b00463443aab6f0dfbdec940466","fd2e09a9cc914ad7986fdf66bc69c4a1","721ba9ab45094523b3881af1d3d7eef4","3af6cb869c0f4ba7a925c8a4bd1b49c7","a5fb37cc8fea4715966ed619b416ef9f","61ef8653bd9848cf986b71b32a9b2fee","585d68b080634af6b82900db7f2bfb3f","acd898044b114cfda79eb2cf37ac4446","7af3031a3afe41c48df83c262098bfc3"]},"executionInfo":{"elapsed":6665,"status":"ok","timestamp":1754981040832,"user":{"displayName":"Grégoire","userId":"01249298206117462766"},"user_tz":-120},"id":"ir7It9rTICQE","outputId":"61b85667-ff0f-4584-b6cd-6ef78f9aff9e"},"outputs":[{"output_type":"stream","name":"stdout","text":["🔍 Vérification des expériences déjà terminées dans 'OC Projet 8'...\n","✅ 39 expériences terminées trouvées:\n"," - FPN_efficientnetb0_finetune_augTrue_prod\n"," - FPN_efficientnetb0_finetune_augTrue_test\n"," - FPN_efficientnetb0_frozen_augFalse_prod\n"," - FPN_efficientnetb0_frozen_augFalse_test\n"," - FPN_efficientnetb0_frozen_augTrue_prod\n"," - FPN_efficientnetb0_frozen_augTrue_test\n"," - FPN_mobilenetv2_frozen_augFalse_prod\n"," - FPN_mobilenetv2_frozen_augFalse_test\n"," - FPN_mobilenetv2_frozen_augTrue_prod\n"," - FPN_mobilenetv2_frozen_augTrue_test\n"," - FPN_resnet34_finetune_augTrue_prod\n"," - FPN_resnet34_finetune_augTrue_test\n"," - FPN_resnet34_frozen_augFalse_prod\n"," - FPN_resnet34_frozen_augFalse_test\n"," - FPN_resnet34_frozen_augTrue_prod\n"," - FPN_resnet34_frozen_augTrue_test\n"," - FPN_vanilla_augFalse_prod\n"," - FPN_vanilla_augFalse_test\n"," - FPN_vanilla_augTrue_prod\n"," - FPN_vanilla_augTrue_test\n"," - U-Net_efficientnetb0_frozen_augFalse_prod\n"," - U-Net_efficientnetb0_frozen_augFalse_test\n"," - U-Net_efficientnetb0_frozen_augTrue_prod\n"," - U-Net_efficientnetb0_frozen_augTrue_test\n"," - U-Net_mobilenetv2_finetune_augTrue_prod\n"," - U-Net_mobilenetv2_finetune_augTrue_test\n"," - U-Net_mobilenetv2_frozen_augFalse_prod\n"," - U-Net_mobilenetv2_frozen_augFalse_test\n"," - U-Net_mobilenetv2_frozen_augTrue_prod\n"," - U-Net_mobilenetv2_frozen_augTrue_test\n"," - U-Net_vanilla_augFalse_prod\n"," - U-Net_vanilla_augFalse_prod_20250730_153443\n"," - U-Net_vanilla_augFalse_test\n"," - U-Net_vanilla_augTrue_prod\n"," - U-Net_vanilla_augTrue_prod_20250730_170602\n"," - U-Net_vanilla_augTrue_test\n"," - exp_001_baseline\n"," - exp_002_no_augmentation\n"," - exp_002_unfrozen\n","🔍 Recherche du meilleur modèle parmi 17 configurations...\n","🏆 MEILLEUR MODÈLE TROUVÉ:\n"," 📊 [15] FPN_efficientnetb0_finetune_augTrue\n"," 🏗️ FPN + efficientnetb0\n"," 🎯 Test Mean IoU: 0.7457\n"," 📈 Test Accuracy: 0.9123\n"," 🆔 Run ID: b6dd1ecdeb3f4b5c899b332f3c948cd7\n","\n"]},{"output_type":"display_data","data":{"text/plain":["Downloading artifacts: 0%| | 0/1 [00:00"],"image/png":"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\n"},"metadata":{}},{"output_type":"stream","name":"stdout","text":["🔢 Matrice de confusion brute (premiers/derniers éléments):\n","============================================================\n","Classes: ['flat', 'human', 'vehicle', 'construction', 'object', 'nature', 'sky', 'void']\n","Matrice (format: lignes=réel, colonnes=prédit):\n","[[9339912 11922 31996 27736 7639 19873 159 96403]\n"," [ 9298 240878 27954 32796 6215 7121 22 7213]\n"," [ 28923 20536 1630480 57085 7436 16871 280 13413]\n"," [ 37107 27385 39861 4762597 51086 145430 14571 61532]\n"," [ 12585 11363 14823 148802 240856 58198 2359 24774]\n"," [ 30020 6106 17487 161547 21272 3703362 10850 23316]\n"," [ 2 2 139 17203 1113 12668 701752 3902]\n"," [ 400815 21207 41511 245295 36253 56041 13049 2267598]]\n"]}],"source":["# Utilisation\n","confusion_result = display_best_confusion_matrix()"]},{"cell_type":"markdown","source":["Affichage de la répartition des modalités dans l'ensemble du dataset d'apprentissage (on passe par le json de matrice de confusion pour gagner du temps)"],"metadata":{"id":"KsVx1dEpBPph"}},{"cell_type":"code","source":["def analyze_class_distribution_from_confusion_matrix(experiment_name=\"OC Projet 8\",\n"," figsize=(14, 8),\n"," color_palette='viridis',\n"," show_values=True):\n"," \"\"\"\n"," Analyse la répartition des classes à partir de la matrice de confusion du meilleur modèle.\n"," Affiche un graphique en barres avec pourcentages (axe gauche) et nombres absolus (axe droite).\n","\n"," Args:\n"," experiment_name (str): Nom de l'expérience MLflow\n"," figsize (tuple): Taille de la figure (largeur, hauteur)\n"," color_palette (str): Palette de couleurs pour les barres\n"," show_values (bool): Afficher les valeurs sur les barres\n","\n"," Returns:\n"," dict: Dictionnaire contenant les statistiques de distribution\n"," \"\"\"\n"," import matplotlib.pyplot as plt\n"," import numpy as np\n"," import json\n"," import tempfile\n"," import seaborn as sns\n","\n"," # Récupérer les informations du meilleur modèle\n"," completed_configs = get_completed_configs_info(experimental_configurations, experiment_name)\n","\n"," if not completed_configs:\n"," print(\"❌ Aucune expérience terminée trouvée\")\n"," return None\n","\n"," # Récupérer le mapping run_name -> run_id depuis MLflow\n"," experiment = mlflow.get_experiment_by_name(experiment_name)\n"," runs_df = mlflow.search_runs(\n"," experiment_ids=[experiment.experiment_id],\n"," filter_string=\"status = 'FINISHED'\",\n"," output_format=\"pandas\"\n"," )\n","\n"," run_name_to_id = {\n"," run.get('tags.mlflow.runName', ''): run['run_id']\n"," for _, run in runs_df.iterrows()\n"," }\n","\n"," # Trouver le meilleur modèle (même logique que display_best_confusion_matrix)\n"," best_result = None\n"," best_iou = -1\n","\n"," for config_idx, config, run_name in completed_configs:\n"," if run_name not in run_name_to_id:\n"," continue\n","\n"," run_id = run_name_to_id[run_name]\n","\n"," try:\n"," run = mlflow.get_run(run_id)\n"," metrics = run.data.metrics\n"," test_mean_iou = metrics.get('test_mean_iou', None)\n","\n"," if test_mean_iou is not None and test_mean_iou > best_iou:\n"," best_iou = test_mean_iou\n"," best_result = {\n"," 'config_idx': config_idx,\n"," 'experiment': config['name'],\n"," 'architecture': config.get('architecture', 'Unknown'),\n"," 'encoder': config.get('encoder', 'None'),\n"," 'test_mean_iou': test_mean_iou,\n"," 'run_name': run_name,\n"," 'run_id': run_id\n"," }\n"," except Exception as e:\n"," continue\n","\n"," if best_result is None:\n"," print(\"❌ Aucun modèle avec test_mean_iou trouvé\")\n"," return None\n","\n"," print(f\"🏆 Analyse de distribution pour le meilleur modèle:\")\n"," print(f\" 📊 [{best_result['config_idx']}] {best_result['experiment']}\")\n"," print(f\" 🎯 Test Mean IoU: {best_result['test_mean_iou']:.4f}\")\n"," print()\n","\n"," # Télécharger et analyser la matrice de confusion\n"," try:\n"," with tempfile.TemporaryDirectory() as temp_dir:\n"," confusion_matrix_path = mlflow.artifacts.download_artifacts(\n"," run_id=best_result['run_id'],\n"," artifact_path=\"confusion_matrix.json\",\n"," dst_path=temp_dir\n"," )\n","\n"," with open(confusion_matrix_path, 'r') as f:\n"," confusion_data = json.load(f)\n","\n"," matrix = np.array(confusion_data['matrix'])\n"," class_names = confusion_data['class_names']\n","\n"," # Calculer la distribution des classes (somme par ligne = nombre total de pixels réels par classe)\n"," class_counts = np.sum(matrix, axis=1) # Somme par ligne pour obtenir le nombre total de chaque classe réelle\n"," total_pixels = np.sum(class_counts)\n"," class_percentages = (class_counts / total_pixels) * 100\n","\n"," # Statistiques détaillées\n"," print(\"📈 RÉPARTITION DES CLASSES DANS L'ÉCHANTILLON:\")\n"," print(\"=\" * 65)\n"," print(f\"{'Classe':<15} {'Nb Pixels':<12} {'Pourcentage':<12} {'Barre':<20}\")\n"," print(\"-\" * 65)\n","\n"," for i, class_name in enumerate(class_names):\n"," bar_length = int(class_percentages[i] / 2) # Barre visuelle\n"," bar = \"█\" * bar_length\n"," print(f\"{class_name:<15} {class_counts[i]:<12,} {class_percentages[i]:<11.2f}% {bar}\")\n","\n"," print(\"-\" * 65)\n"," print(f\"{'TOTAL':<15} {total_pixels:<12,} {'100.00%':<12}\")\n"," print()\n","\n"," # Créer le graphique\n"," fig, ax1 = plt.subplots(figsize=figsize)\n","\n"," # Utiliser une palette de couleurs\n"," colors = plt.cm.get_cmap(color_palette)(np.linspace(0, 1, len(class_names)))\n","\n"," # Barres avec les pourcentages (axe gauche)\n"," bars = ax1.bar(range(len(class_names)), class_percentages,\n"," color=colors, alpha=0.8, edgecolor='black', linewidth=1)\n","\n"," # Configuration axe gauche (pourcentages)\n"," ax1.set_xlabel('Classes', fontsize=12, fontweight='bold')\n"," ax1.set_ylabel('Pourcentage (%)', fontsize=12, fontweight='bold', color='blue')\n"," ax1.tick_params(axis='y', labelcolor='blue')\n"," ax1.set_xticks(range(len(class_names)))\n"," ax1.set_xticklabels(class_names, rotation=45, ha='right')\n"," ax1.grid(True, alpha=0.3, axis='y')\n","\n"," # Créer un second axe y pour les nombres absolus\n"," ax2 = ax1.twinx()\n"," ax2.set_ylabel('Nombre de Pixels', fontsize=12, fontweight='bold', color='red')\n"," ax2.tick_params(axis='y', labelcolor='red')\n","\n"," # Synchroniser les échelles\n"," ax2.set_ylim(0, np.max(class_counts) * 1.1)\n"," ax1.set_ylim(0, np.max(class_percentages) * 1.1)\n","\n"," # Ajouter les valeurs sur les barres si demandé\n"," if show_values:\n"," for i, (bar, count, pct) in enumerate(zip(bars, class_counts, class_percentages)):\n"," height = bar.get_height()\n"," # Pourcentage au-dessus de la barre\n"," ax1.text(bar.get_x() + bar.get_width()/2., height + 0.5,\n"," f'{pct:.1f}%', ha='center', va='bottom', fontweight='bold', fontsize=10)\n"," # Nombre absolu à l'intérieur de la barre\n"," ax1.text(bar.get_x() + bar.get_width()/2., height/2,\n"," f'{count:,}', ha='center', va='center',\n"," color='white', fontweight='bold', fontsize=9)\n","\n"," # Titre et mise en forme\n"," plt.title(f'Répartition des Classes par Pixels\\n'\n"," f'Total: {total_pixels:,} pixels | Mean IoU: {best_result[\"test_mean_iou\"]:.4f}',\n"," fontsize=14, fontweight='bold', pad=20)\n","\n"," plt.tight_layout()\n"," plt.show()\n","\n"," # Statistiques supplémentaires\n"," print(\"📊 STATISTIQUES COMPLÉMENTAIRES:\")\n"," print(\"-\" * 40)\n"," print(f\" 🔢 Nombre total de pixels: {total_pixels:,}\")\n"," print(f\" 📏 Nombre de classes: {len(class_names)}\")\n"," print(f\" 📈 Classe la plus représentée: {class_names[np.argmax(class_counts)]} ({np.max(class_percentages):.2f}%)\")\n"," print(f\" 📉 Classe la moins représentée: {class_names[np.argmin(class_counts)]} ({np.min(class_percentages):.2f}%)\")\n"," print(f\" ⚖️ Ratio max/min: {np.max(class_counts)/np.min(class_counts):.2f}\")\n","\n"," # Calcul de l'équilibrage des classes\n"," ideal_percentage = 100 / len(class_names)\n"," imbalance_scores = np.abs(class_percentages - ideal_percentage)\n"," avg_imbalance = np.mean(imbalance_scores)\n"," print(f\" ⚖️ Score de déséquilibre moyen: {avg_imbalance:.2f}% (0 = parfait équilibre)\")\n"," print()\n","\n"," return {\n"," 'class_names': class_names,\n"," 'class_counts': class_counts,\n"," 'class_percentages': class_percentages,\n"," 'total_pixels': total_pixels,\n"," 'best_model_info': best_result,\n"," 'most_common_class': class_names[np.argmax(class_counts)],\n"," 'least_common_class': class_names[np.argmin(class_counts)],\n"," 'imbalance_ratio': np.max(class_counts)/np.min(class_counts),\n"," 'avg_imbalance_score': avg_imbalance\n"," }\n","\n"," except Exception as e:\n"," print(f\"❌ Erreur lors de l'analyse de distribution: {e}\")\n"," print(f\" Run ID: {best_result['run_id']}\")\n"," return None\n","\n","\n","def compare_class_distributions(experiment_name=\"OC Projet 8\", top_n=3):\n"," \"\"\"\n"," Compare la répartition des classes entre les N meilleurs modèles.\n","\n"," Args:\n"," experiment_name (str): Nom de l'expérience MLflow\n"," top_n (int): Nombre de meilleurs modèles à comparer\n"," \"\"\"\n"," import matplotlib.pyplot as plt\n"," import numpy as np\n"," import json\n"," import tempfile\n"," import seaborn as sns\n","\n"," # Récupérer et trier les modèles par performance\n"," completed_configs = get_completed_configs_info(experimental_configurations, experiment_name)\n","\n"," if not completed_configs:\n"," print(\"❌ Aucune expérience terminée trouvée\")\n"," return None\n","\n"," experiment = mlflow.get_experiment_by_name(experiment_name)\n"," runs_df = mlflow.search_runs(\n"," experiment_ids=[experiment.experiment_id],\n"," filter_string=\"status = 'FINISHED'\",\n"," output_format=\"pandas\"\n"," )\n","\n"," run_name_to_id = {\n"," run.get('tags.mlflow.runName', ''): run['run_id']\n"," for _, run in runs_df.iterrows()\n"," }\n","\n"," # Collecter les performances de tous les modèles\n"," models_performance = []\n","\n"," for config_idx, config, run_name in completed_configs:\n"," if run_name not in run_name_to_id:\n"," continue\n","\n"," run_id = run_name_to_id[run_name]\n","\n"," try:\n"," run = mlflow.get_run(run_id)\n"," metrics = run.data.metrics\n"," test_mean_iou = metrics.get('test_mean_iou', None)\n","\n"," if test_mean_iou is not None:\n"," models_performance.append({\n"," 'config_idx': config_idx,\n"," 'experiment': config['name'],\n"," 'test_mean_iou': test_mean_iou,\n"," 'run_name': run_name,\n"," 'run_id': run_id\n"," })\n"," except Exception as e:\n"," continue\n","\n"," # Trier par performance et prendre les N meilleurs\n"," models_performance.sort(key=lambda x: x['test_mean_iou'], reverse=True)\n"," top_models = models_performance[:top_n]\n","\n"," print(f\"🏆 COMPARAISON DES {len(top_models)} MEILLEURS MODÈLES:\")\n"," print(\"=\" * 60)\n","\n"," # Analyser chaque modèle\n"," all_distributions = []\n"," class_names = None\n","\n"," for i, model in enumerate(top_models):\n"," print(f\"📊 Modèle #{i+1}: {model['experiment']} (IoU: {model['test_mean_iou']:.4f})\")\n","\n"," try:\n"," with tempfile.TemporaryDirectory() as temp_dir:\n"," confusion_matrix_path = mlflow.artifacts.download_artifacts(\n"," run_id=model['run_id'],\n"," artifact_path=\"confusion_matrix.json\",\n"," dst_path=temp_dir\n"," )\n","\n"," with open(confusion_matrix_path, 'r') as f:\n"," confusion_data = json.load(f)\n","\n"," matrix = np.array(confusion_data['matrix'])\n"," if class_names is None:\n"," class_names = confusion_data['class_names']\n","\n"," class_counts = np.sum(matrix, axis=1)\n"," total_pixels = np.sum(class_counts)\n"," class_percentages = (class_counts / total_pixels) * 100\n","\n"," all_distributions.append({\n"," 'model_name': f\"Modèle {i+1}\",\n"," 'percentages': class_percentages,\n"," 'counts': class_counts,\n"," 'iou': model['test_mean_iou']\n"," })\n","\n"," except Exception as e:\n"," print(f\" ⚠️ Erreur: {e}\")\n"," continue\n","\n"," if not all_distributions:\n"," print(\"❌ Impossible de récupérer les données de distribution\")\n"," return None\n","\n"," # Créer le graphique de comparaison\n"," fig, (ax1, ax2) = plt.subplots(1, 2, figsize=(16, 8))\n","\n"," # Graphique 1: Pourcentages\n"," x_pos = np.arange(len(class_names))\n"," bar_width = 0.8 / len(all_distributions)\n","\n"," for i, dist in enumerate(all_distributions):\n"," ax1.bar(x_pos + i * bar_width, dist['percentages'],\n"," bar_width, label=f\"{dist['model_name']} (IoU: {dist['iou']:.3f})\",\n"," alpha=0.8)\n","\n"," ax1.set_xlabel('Classes')\n"," ax1.set_ylabel('Pourcentage (%)')\n"," ax1.set_title('Comparaison des Distributions par Classe (%)')\n"," ax1.set_xticks(x_pos + bar_width * (len(all_distributions) - 1) / 2)\n"," ax1.set_xticklabels(class_names, rotation=45, ha='right')\n"," ax1.legend()\n"," ax1.grid(True, alpha=0.3)\n","\n"," # Graphique 2: Heatmap des différences relatives au meilleur modèle\n"," reference_dist = all_distributions[0]['percentages'] # Le meilleur modèle comme référence\n"," diff_matrix = []\n","\n"," for dist in all_distributions:\n"," relative_diff = ((dist['percentages'] - reference_dist) / reference_dist) * 100\n"," diff_matrix.append(relative_diff)\n","\n"," sns.heatmap(diff_matrix,\n"," xticklabels=class_names,\n"," yticklabels=[dist['model_name'] for dist in all_distributions],\n"," cmap='RdBu_r', center=0, annot=True, fmt='.1f',\n"," cbar_kws={'label': 'Différence relative (%)'},\n"," ax=ax2)\n","\n"," ax2.set_title(f'Différences Relatives au Meilleur Modèle\\n(Référence: {all_distributions[0][\"model_name\"]})')\n"," ax2.set_xlabel('Classes')\n","\n"," plt.tight_layout()\n"," plt.show()\n","\n"," return {\n"," 'top_models': top_models,\n"," 'distributions': all_distributions,\n"," 'class_names': class_names\n"," }"],"metadata":{"id":"LXs3Sr5pA6uJ","executionInfo":{"status":"ok","timestamp":1754981151442,"user_tz":-120,"elapsed":164,"user":{"displayName":"Grégoire","userId":"01249298206117462766"}}},"execution_count":37,"outputs":[]},{"cell_type":"code","source":["# Analyse simple du meilleur modèle\n","result = analyze_class_distribution_from_confusion_matrix(\"OC Projet 8\")\n","\n","# Avec personnalisation\n","result = analyze_class_distribution_from_confusion_matrix(\n"," experiment_name=\"OC Projet 8\",\n"," figsize=(16, 10),\n"," color_palette='plasma',\n"," show_values=True\n",")\n","\n","# Comparaison entre plusieurs modèles\n","comparison = compare_class_distributions(\"OC Projet 8\", top_n=3)"],"metadata":{"colab":{"base_uri":"https://localhost:8080/","height":1000,"referenced_widgets":["0814405d6086476bbb2ce5ff31ee24fe","1288de0fadb54c03bb1a62c8f9a54076","c31c3ddb9fc743a1b9b3f88a1415819f","8072bb3d4816439ba2e706ec1800f602","745627c8443b43a8be3600bac3a03f75","39e52ed0f4ab4a00b0b8b8dd79227d09","28d74235f7af44d58fcdd73540130395","af907ef5c98f45e7bcdb9dfb4b9c4e19","f16b43ee72db482db365dc19b34279d2","afb8e1ef0c7642f18b0a55061ced6ba0","86223e1bc4774d59b854e28e479f5fa6","ea57b0429a394fdca0b9d1c199705f62","285d0104a49749f0914caab4633ba95f","b1954114b94442a1afef5fef5bd197e1","7703627bc5f84c01adf142b882dea1a0","b5de34b2a7114cb78c443bc9c57dd913","0c7daff8beaa46918b2db8a2e598c55d","b61fa6b5f26e418ca659e96b46aeec7b","2dea319059604490bc4dc6ccdf32f90c","470f3d2cd2c34156ac4660ff19ad0cd5","b2805d79bf224a259bcf31504e6a98a0","1d0403d7a48f4f78a5ed385e26141102","3e1a19271ccc42e1bbcb78b0a3bed498","fb03d61318424159a68af92d7b5905fd","b39d3dcf293e4fbd86c04ef6064c684d","6a5fa67683cb430a8f82ccac7ed8df1d","0bf8a446cb6a4314a61a007f6532fd3a","db5a00931ef44d07a5f784aa124b3925","983f9bc3c7164abab23c6fcde4ba1bd7","fcb838679d264e51bff81d1fa8a0a110","e2c87ebdaf5d4f74a445a3892e3fd7ad","6b7a0d91fc52443cbf807d2bfcd29de2","34c6f6b3248e496796002094701b6e14","e0b2701aedc84593a9a48a4a05057b42","5020a5cc9d3c45c09b16915d8f5aca8e","4a91a17d98624c16982a073a53ef3acd","7049cef08ba04207b58db406da46e655","8354fa35766a402ea853f31210188fa5","18e09f691f8c48bda8c80dbd8d3cdcbf","63e8b18b98f74a6b8ecc4c37582ddc34","b3278303d86446d8a9321d401667adc8","6752fb87421d49989fe3c5716770c983","1e3af0148879401da1e96b5fe593cf1b","b7399763b1ca42799cad80619bb34756","1e46e810a51146c191c083337215012c","5febffdeb3f344b782112170fe40433e","5260f52807fa4b979070284a3738759f","e8338bea77b34ade8d3461ad408470a8","2b490d66948f448c81896b9c56d44c2a","43e14174355b4441803a94e9306e45df","6a088a189bc94acf87aa1099c2bec9fd","00577047db6d4fe3b949a33e87e60e4b","823e48aabb094171a043e70e1b817683","12c8c85d30694229b21951e6415e1425","b427a4205526434c9e1649cb0a631b13"]},"id":"VoTFNfGhBY8x","executionInfo":{"status":"ok","timestamp":1754981172370,"user_tz":-120,"elapsed":19492,"user":{"displayName":"Grégoire","userId":"01249298206117462766"}},"outputId":"e41a5021-52f8-4718-a0fb-80d9afaa0c82"},"execution_count":38,"outputs":[{"output_type":"stream","name":"stdout","text":["🔍 Vérification des expériences déjà terminées dans 'OC Projet 8'...\n","✅ 39 expériences terminées trouvées:\n"," - FPN_efficientnetb0_finetune_augTrue_prod\n"," - FPN_efficientnetb0_finetune_augTrue_test\n"," - FPN_efficientnetb0_frozen_augFalse_prod\n"," - FPN_efficientnetb0_frozen_augFalse_test\n"," - FPN_efficientnetb0_frozen_augTrue_prod\n"," - FPN_efficientnetb0_frozen_augTrue_test\n"," - FPN_mobilenetv2_frozen_augFalse_prod\n"," - FPN_mobilenetv2_frozen_augFalse_test\n"," - FPN_mobilenetv2_frozen_augTrue_prod\n"," - FPN_mobilenetv2_frozen_augTrue_test\n"," - FPN_resnet34_finetune_augTrue_prod\n"," - FPN_resnet34_finetune_augTrue_test\n"," - FPN_resnet34_frozen_augFalse_prod\n"," - FPN_resnet34_frozen_augFalse_test\n"," - FPN_resnet34_frozen_augTrue_prod\n"," - FPN_resnet34_frozen_augTrue_test\n"," - FPN_vanilla_augFalse_prod\n"," - FPN_vanilla_augFalse_test\n"," - FPN_vanilla_augTrue_prod\n"," - FPN_vanilla_augTrue_test\n"," - U-Net_efficientnetb0_frozen_augFalse_prod\n"," - U-Net_efficientnetb0_frozen_augFalse_test\n"," - U-Net_efficientnetb0_frozen_augTrue_prod\n"," - U-Net_efficientnetb0_frozen_augTrue_test\n"," - U-Net_mobilenetv2_finetune_augTrue_prod\n"," - U-Net_mobilenetv2_finetune_augTrue_test\n"," - U-Net_mobilenetv2_frozen_augFalse_prod\n"," - U-Net_mobilenetv2_frozen_augFalse_test\n"," - U-Net_mobilenetv2_frozen_augTrue_prod\n"," - U-Net_mobilenetv2_frozen_augTrue_test\n"," - U-Net_vanilla_augFalse_prod\n"," - U-Net_vanilla_augFalse_prod_20250730_153443\n"," - U-Net_vanilla_augFalse_test\n"," - U-Net_vanilla_augTrue_prod\n"," - U-Net_vanilla_augTrue_prod_20250730_170602\n"," - U-Net_vanilla_augTrue_test\n"," - exp_001_baseline\n"," - exp_002_no_augmentation\n"," - exp_002_unfrozen\n","🏆 Analyse de distribution pour le meilleur modèle:\n"," 📊 [15] FPN_efficientnetb0_finetune_augTrue\n"," 🎯 Test Mean IoU: 0.7457\n","\n"]},{"output_type":"display_data","data":{"text/plain":["Downloading artifacts: 0%| | 0/1 [00:00"],"image/png":"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\n"},"metadata":{}},{"output_type":"stream","name":"stdout","text":["📊 STATISTIQUES COMPLÉMENTAIRES:\n","----------------------------------------\n"," 🔢 Nombre total de pixels: 25,088,000\n"," 📏 Nombre de classes: 8\n"," 📈 Classe la plus représentée: flat (38.01%)\n"," 📉 Classe la moins représentée: human (1.32%)\n"," ⚖️ Ratio max/min: 28.77\n"," ⚖️ Score de déséquilibre moyen: 9.21% (0 = parfait équilibre)\n","\n","🔍 Vérification des expériences déjà terminées dans 'OC Projet 8'...\n","✅ 39 expériences terminées trouvées:\n"," - FPN_efficientnetb0_finetune_augTrue_prod\n"," - FPN_efficientnetb0_finetune_augTrue_test\n"," - FPN_efficientnetb0_frozen_augFalse_prod\n"," - FPN_efficientnetb0_frozen_augFalse_test\n"," - FPN_efficientnetb0_frozen_augTrue_prod\n"," - FPN_efficientnetb0_frozen_augTrue_test\n"," - FPN_mobilenetv2_frozen_augFalse_prod\n"," - FPN_mobilenetv2_frozen_augFalse_test\n"," - FPN_mobilenetv2_frozen_augTrue_prod\n"," - FPN_mobilenetv2_frozen_augTrue_test\n"," - FPN_resnet34_finetune_augTrue_prod\n"," - FPN_resnet34_finetune_augTrue_test\n"," - FPN_resnet34_frozen_augFalse_prod\n"," - FPN_resnet34_frozen_augFalse_test\n"," - FPN_resnet34_frozen_augTrue_prod\n"," - FPN_resnet34_frozen_augTrue_test\n"," - FPN_vanilla_augFalse_prod\n"," - FPN_vanilla_augFalse_test\n"," - FPN_vanilla_augTrue_prod\n"," - FPN_vanilla_augTrue_test\n"," - U-Net_efficientnetb0_frozen_augFalse_prod\n"," - U-Net_efficientnetb0_frozen_augFalse_test\n"," - U-Net_efficientnetb0_frozen_augTrue_prod\n"," - U-Net_efficientnetb0_frozen_augTrue_test\n"," - U-Net_mobilenetv2_finetune_augTrue_prod\n"," - U-Net_mobilenetv2_finetune_augTrue_test\n"," - U-Net_mobilenetv2_frozen_augFalse_prod\n"," - U-Net_mobilenetv2_frozen_augFalse_test\n"," - U-Net_mobilenetv2_frozen_augTrue_prod\n"," - U-Net_mobilenetv2_frozen_augTrue_test\n"," - U-Net_vanilla_augFalse_prod\n"," - U-Net_vanilla_augFalse_prod_20250730_153443\n"," - U-Net_vanilla_augFalse_test\n"," - U-Net_vanilla_augTrue_prod\n"," - U-Net_vanilla_augTrue_prod_20250730_170602\n"," - U-Net_vanilla_augTrue_test\n"," - exp_001_baseline\n"," - exp_002_no_augmentation\n"," - exp_002_unfrozen\n","🏆 Analyse de distribution pour le meilleur modèle:\n"," 📊 [15] FPN_efficientnetb0_finetune_augTrue\n"," 🎯 Test Mean IoU: 0.7457\n","\n"]},{"output_type":"display_data","data":{"text/plain":["Downloading artifacts: 0%| | 0/1 [00:00"],"image/png":"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\n"},"metadata":{}},{"output_type":"stream","name":"stdout","text":["📊 STATISTIQUES COMPLÉMENTAIRES:\n","----------------------------------------\n"," 🔢 Nombre total de pixels: 25,088,000\n"," 📏 Nombre de classes: 8\n"," 📈 Classe la plus représentée: flat (38.01%)\n"," 📉 Classe la moins représentée: human (1.32%)\n"," ⚖️ Ratio max/min: 28.77\n"," ⚖️ Score de déséquilibre moyen: 9.21% (0 = parfait équilibre)\n","\n","🔍 Vérification des expériences déjà terminées dans 'OC Projet 8'...\n","✅ 39 expériences terminées trouvées:\n"," - FPN_efficientnetb0_finetune_augTrue_prod\n"," - FPN_efficientnetb0_finetune_augTrue_test\n"," - FPN_efficientnetb0_frozen_augFalse_prod\n"," - FPN_efficientnetb0_frozen_augFalse_test\n"," - FPN_efficientnetb0_frozen_augTrue_prod\n"," - FPN_efficientnetb0_frozen_augTrue_test\n"," - FPN_mobilenetv2_frozen_augFalse_prod\n"," - FPN_mobilenetv2_frozen_augFalse_test\n"," - FPN_mobilenetv2_frozen_augTrue_prod\n"," - FPN_mobilenetv2_frozen_augTrue_test\n"," - FPN_resnet34_finetune_augTrue_prod\n"," - FPN_resnet34_finetune_augTrue_test\n"," - FPN_resnet34_frozen_augFalse_prod\n"," - FPN_resnet34_frozen_augFalse_test\n"," - FPN_resnet34_frozen_augTrue_prod\n"," - FPN_resnet34_frozen_augTrue_test\n"," - FPN_vanilla_augFalse_prod\n"," - FPN_vanilla_augFalse_test\n"," - FPN_vanilla_augTrue_prod\n"," - FPN_vanilla_augTrue_test\n"," - U-Net_efficientnetb0_frozen_augFalse_prod\n"," - U-Net_efficientnetb0_frozen_augFalse_test\n"," - U-Net_efficientnetb0_frozen_augTrue_prod\n"," - U-Net_efficientnetb0_frozen_augTrue_test\n"," - U-Net_mobilenetv2_finetune_augTrue_prod\n"," - U-Net_mobilenetv2_finetune_augTrue_test\n"," - U-Net_mobilenetv2_frozen_augFalse_prod\n"," - U-Net_mobilenetv2_frozen_augFalse_test\n"," - U-Net_mobilenetv2_frozen_augTrue_prod\n"," - U-Net_mobilenetv2_frozen_augTrue_test\n"," - U-Net_vanilla_augFalse_prod\n"," - U-Net_vanilla_augFalse_prod_20250730_153443\n"," - U-Net_vanilla_augFalse_test\n"," - U-Net_vanilla_augTrue_prod\n"," - U-Net_vanilla_augTrue_prod_20250730_170602\n"," - U-Net_vanilla_augTrue_test\n"," - exp_001_baseline\n"," - exp_002_no_augmentation\n"," - exp_002_unfrozen\n","🏆 COMPARAISON DES 3 MEILLEURS MODÈLES:\n","============================================================\n","📊 Modèle #1: FPN_efficientnetb0_finetune_augTrue (IoU: 0.7457)\n"]},{"output_type":"display_data","data":{"text/plain":["Downloading artifacts: 0%| | 0/1 [00:00"],"image/png":"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\n"},"metadata":{}}]},{"cell_type":"markdown","metadata":{"id":"8-vXPIYNSALV"},"source":["### Mise en production meilleur modèle"]},{"cell_type":"code","source":["def register_best_model(experiment_name=\"OC Projet 8\", model_name=\"StreetsSegmentation\",\n"," force=False, specific_run_id=None):\n"," \"\"\"\n"," Enregistre le meilleur modèle (selon test_mean_iou) OU un modèle spécifique dans le Registry MLflow.\n","\n"," Args:\n"," experiment_name (str): Nom de l'expérience MLflow\n"," model_name (str): Nom du modèle dans le Registry\n"," force (bool): Si True, force la mise en production même si le score est moins bon\n"," specific_run_id (str): Si fourni, enregistre ce run spécifique au lieu du meilleur\n"," \"\"\"\n"," print(f\"\\n{'='*60}\")\n"," if specific_run_id:\n"," print(f\"🎯 Enregistrement d'un modèle SPÉCIFIQUE dans le Registry MLflow\")\n"," print(f\"🆔 Run ID spécifique: {specific_run_id}\")\n"," else:\n"," print(f\"🏆 Enregistrement du MEILLEUR modèle dans le Registry MLflow\")\n"," print(f\"🔬 Expérience MLflow: {experiment_name}\")\n"," print(f\"📦 Nom du modèle Registry: {model_name}\")\n"," if force:\n"," print(f\"⚡ Mode FORCE activé - Mise en prod forcée\")\n"," print(f\"{'='*60}\")\n","\n"," # 1. Si un run_id spécifique est fourni, l'utiliser directement\n"," if specific_run_id:\n"," print(f\"🎯 Mode spécifique activé - utilisation du run {specific_run_id}\")\n","\n"," try:\n"," # Vérifier que le run existe et récupérer ses métriques\n"," run = mlflow.get_run(specific_run_id)\n"," metrics = run.data.metrics\n"," params = run.data.params\n","\n"," test_mean_iou = metrics.get('test_mean_iou', None)\n"," test_accuracy = metrics.get('test_sparse_categorical_accuracy', None)\n"," run_name = run.data.tags.get('mlflow.runName', 'Unknown')\n","\n"," if test_mean_iou is None:\n"," print(\"⚠️ Le run spécifique n'a pas de métrique test_mean_iou\")\n"," if not force:\n"," print(\"❌ Utilisez force=True pour forcer l'enregistrement sans métriques\")\n"," return None\n","\n"," selected_result = {\n"," 'run_id': specific_run_id,\n"," 'test_mean_iou': test_mean_iou or -1.0,\n"," 'test_accuracy': test_accuracy,\n"," 'experiment_name': run_name,\n"," 'selection_mode': 'specific'\n"," }\n","\n"," print(f\"✅ Run spécifique trouvé:\")\n"," print(f\" • Run Name: {run_name}\")\n"," print(f\" • Run ID: {specific_run_id}\")\n"," if test_mean_iou:\n"," print(f\" • Test Mean IoU: {test_mean_iou:.4f}\")\n"," if test_accuracy:\n"," print(f\" • Test Accuracy: {test_accuracy:.4f}\")\n","\n"," except Exception as e:\n"," print(f\"❌ Erreur lors de la récupération du run spécifique {specific_run_id}: {e}\")\n"," return None\n","\n"," else:\n"," # 2. Mode automatique : trouver le meilleur modèle (logique existante)\n"," print(\"🔍 Mode automatique - recherche du meilleur modèle...\")\n","\n"," # Récupérer les configurations terminées (réutilise la logique existante)\n"," completed_configs = get_completed_configs_info(experimental_configurations, experiment_name)\n","\n"," if not completed_configs:\n"," print(\"❌ Aucune expérience terminée trouvée\")\n"," return None\n","\n"," # Récupérer les métriques MLflow pour chaque run\n"," experiment = mlflow.get_experiment_by_name(experiment_name)\n"," runs_df = mlflow.search_runs(\n"," experiment_ids=[experiment.experiment_id],\n"," filter_string=\"status = 'FINISHED'\",\n"," output_format=\"pandas\"\n"," )\n","\n"," run_name_to_id = {\n"," run.get('tags.mlflow.runName', ''): run['run_id']\n"," for _, run in runs_df.iterrows()\n"," }\n","\n"," print(f\"📊 Analyse de {len(completed_configs)} configurations terminées\")\n","\n"," # Collecter les résultats avec métriques\n"," results = []\n"," for config_idx, config, run_name in completed_configs:\n"," if run_name not in run_name_to_id:\n"," continue\n","\n"," run_id = run_name_to_id[run_name]\n","\n"," try:\n"," # Récupérer les métriques directement depuis MLflow\n"," run = mlflow.get_run(run_id)\n"," metrics = run.data.metrics\n","\n"," test_mean_iou = metrics.get('test_mean_iou', None)\n","\n"," if test_mean_iou is not None:\n"," results.append({\n"," 'config_idx': config_idx,\n"," 'config': config,\n"," 'run_name': run_name,\n"," 'run_id': run_id,\n"," 'test_mean_iou': test_mean_iou,\n"," 'experiment_name': config['name']\n"," })\n"," print(f\" ✅ [{config_idx}] {config['name']}: IoU={test_mean_iou:.4f}\")\n"," else:\n"," print(f\" ⚠️ [{config_idx}] {config['name']}: Pas de test_mean_iou\")\n","\n"," except Exception as e:\n"," print(f\" ❌ [{config_idx}] {config['name']}: Erreur - {e}\")\n"," continue\n","\n"," if not results:\n"," print(\"❌ Aucune expérience avec test_mean_iou trouvée\")\n"," return None\n","\n"," # Trouver le meilleur modèle\n"," best_result = max(results, key=lambda x: x['test_mean_iou'])\n","\n"," selected_result = {\n"," 'run_id': best_result['run_id'],\n"," 'test_mean_iou': best_result['test_mean_iou'],\n"," 'experiment_name': best_result['experiment_name'],\n"," 'selection_mode': 'automatic'\n"," }\n","\n"," print(f\"🏆 Meilleur modèle identifié:\")\n"," print(f\" • Expérience: {best_result['experiment_name']}\")\n"," print(f\" • Run ID: {best_result['run_id']}\")\n"," print(f\" • Test Mean IoU: {best_result['test_mean_iou']:.4f}\")\n","\n"," # 3. Vérifier si le modèle existe dans le Registry (reste identique)\n"," client = mlflow.client.MlflowClient()\n"," current_best_iou = -1.0\n"," model_exists = False\n"," latest_version = None\n","\n"," try:\n"," try:\n"," registered_model = client.get_registered_model(model_name)\n"," model_exists = True\n"," print(f\"\\n📦 Modèle '{model_name}' trouvé dans le Registry\")\n"," except mlflow.exceptions.RestException:\n"," model_exists = False\n"," print(f\"\\n📦 Modèle '{model_name}' n'existe pas encore dans le Registry\")\n","\n"," if model_exists:\n"," model_versions = client.search_model_versions(f\"name='{model_name}'\")\n"," if model_versions:\n"," latest_version = max(model_versions, key=lambda v: int(v.version))\n"," current_run = mlflow.get_run(latest_version.run_id)\n"," current_best_iou = current_run.data.metrics.get('test_mean_iou', -1.0)\n","\n"," print(f\" • Version la plus récente: {latest_version.version}\")\n"," print(f\" • Stage: {latest_version.current_stage}\")\n"," print(f\" • Test Mean IoU: {current_best_iou:.4f}\")\n","\n"," except Exception as e:\n"," print(f\"⚠️ Erreur lors de la vérification du Registry: {e}\")\n","\n"," # 4. Logique de comparaison adaptée\n"," if specific_run_id:\n"," # En mode spécifique, on enregistre toujours (sauf si force=False et pas de métriques)\n"," should_register = True\n"," if selected_result['test_mean_iou'] == -1.0 and not force:\n"," should_register = False\n"," print(\"❌ Le run spécifique n'a pas de métriques et force=False\")\n"," else:\n"," # En mode automatique, comparaison habituelle\n"," should_register = selected_result['test_mean_iou'] > current_best_iou or force\n","\n"," # 5. Enregistrement (logique existante adaptée)\n"," if should_register:\n"," if specific_run_id:\n"," print(f\"\\n🎯 Enregistrement du modèle spécifique...\")\n"," if selected_result['test_mean_iou'] > current_best_iou:\n"," action_msg = f\"Le modèle spécifique ({selected_result['test_mean_iou']:.4f}) est meilleur que l'actuel ({current_best_iou:.4f})\"\n"," elif force:\n"," action_msg = f\"Enregistrement FORCÉ du modèle spécifique\"\n"," else:\n"," action_msg = f\"Enregistrement du modèle spécifique\"\n"," print(f\" {action_msg}\")\n"," else:\n"," if not model_exists:\n"," print(f\"\\n🆕 Création du modèle '{model_name}' dans le Registry...\")\n"," action_msg = \"Création et enregistrement du premier modèle\"\n"," elif force and selected_result['test_mean_iou'] <= current_best_iou:\n"," print(f\"\\n⚡ FORCE MODE: Enregistrement forcé du modèle\")\n"," print(f\" Nouveau: {selected_result['test_mean_iou']:.4f} vs Actuel: {current_best_iou:.4f}\")\n"," action_msg = \"Enregistrement forcé de la nouvelle version\"\n"," else:\n"," print(f\"\\n🚀 Le nouveau modèle ({selected_result['test_mean_iou']:.4f}) est meilleur que l'actuel ({current_best_iou:.4f})\")\n"," action_msg = \"Enregistrement de la nouvelle version\"\n","\n"," print(f\" {action_msg}...\")\n","\n"," try:\n"," # Si le modèle n'existe pas, le créer\n"," if not model_exists:\n"," client.create_registered_model(\n"," name=model_name,\n"," description=f\"Modèle de segmentation des rues - Performance: {selected_result['test_mean_iou']:.4f}\"\n"," )\n"," print(f\" ✅ Modèle '{model_name}' créé dans le Registry\")\n","\n"," # Vérifier si une version existe déjà pour ce run spécifique\n"," existing_versions = [v for v in client.search_model_versions(f\"name='{model_name}'\")\n"," if v.run_id == selected_result['run_id']]\n","\n"," if existing_versions:\n"," model_version = existing_versions[0]\n"," print(f\" ℹ️ Version existante trouvée pour ce run: {model_version.version}\")\n"," else:\n"," print(\" 📝 Enregistrement d'une nouvelle version...\")\n","\n"," # Chercher l'artefact du modèle\n"," artifacts = client.list_artifacts(selected_result['run_id'])\n"," model_artifact_path = None\n","\n"," for artifact in artifacts:\n"," if any(pattern in artifact.path.lower() for pattern in\n"," ['model', 'best_model', '.keras', '.h5', '.pkl']):\n"," model_artifact_path = artifact.path\n"," break\n","\n"," if model_artifact_path is None:\n"," print(\" ❌ Aucun artefact de modèle trouvé dans le run\")\n"," return None\n","\n"," print(f\" 📁 Artefact trouvé: {model_artifact_path}\")\n","\n"," model_version = client.create_model_version(\n"," name=model_name,\n"," source=f\"runs:/{selected_result['run_id']}/{model_artifact_path}\",\n"," run_id=selected_result['run_id'],\n"," description=f\"Model from {selected_result['experiment_name']} - IoU: {selected_result['test_mean_iou']:.4f} - Mode: {selected_result['selection_mode']}\"\n"," )\n","\n"," # Ajouter les tags\n"," print(\" 🏷️ Ajout des tags de métadonnées...\")\n"," tags_to_add = {\n"," \"source_run_id\": selected_result['run_id'],\n"," \"test_mean_iou\": str(selected_result['test_mean_iou']),\n"," \"model_name\": model_name,\n"," \"source_experiment\": selected_result['experiment_name'],\n"," \"selection_mode\": selected_result['selection_mode'],\n"," \"registration_mode\": \"forced\" if force and selected_result['test_mean_iou'] <= current_best_iou else \"automatic\"\n"," }\n","\n"," for tag_key, tag_value in tags_to_add.items():\n"," client.set_model_version_tag(\n"," name=model_name,\n"," version=model_version.version,\n"," key=tag_key,\n"," value=tag_value\n"," )\n","\n"," # Promouvoir en Production\n"," print(f\" 🚀 Promotion de la version {model_version.version} en Production...\")\n"," client.transition_model_version_stage(\n"," name=model_name,\n"," version=model_version.version,\n"," stage=\"Production\",\n"," archive_existing_versions=True\n"," )\n","\n"," print(f\"✅ Modèle enregistré avec succès !\")\n"," print(f\" • Nom: {model_name}\")\n"," print(f\" • Version: {model_version.version}\")\n"," print(f\" • Stage: Production\")\n"," print(f\" • Test Mean IoU: {selected_result['test_mean_iou']:.4f}\")\n"," print(f\" • Mode de sélection: {selected_result['selection_mode']}\")\n","\n"," return {\n"," 'model_name': model_name,\n"," 'version': model_version.version,\n"," 'run_id': selected_result['run_id'],\n"," 'test_mean_iou': selected_result['test_mean_iou'],\n"," 'selection_mode': selected_result['selection_mode'],\n"," 'action': 'created' if not model_exists else 'updated',\n"," 'tags': tags_to_add\n"," }\n","\n"," except Exception as e:\n"," print(f\"❌ Erreur lors de l'enregistrement: {e}\")\n"," return None\n"," else:\n"," print(f\"\\n➡️ Aucun enregistrement effectué\")\n"," if specific_run_id:\n"," print(\" Le run spécifique n'a pas de métriques suffisantes\")\n"," print(\" 💡 Utilisez force=True pour forcer l'enregistrement\")\n"," else:\n"," print(f\" La version actuelle ({current_best_iou:.4f}) est déjà optimale\")\n"," print(\" 💡 Utilisez force=True pour forcer la mise en production\")\n","\n"," return {\n"," 'model_name': model_name,\n"," 'action': 'no_update',\n"," 'reason': 'Insufficient metrics' if specific_run_id else 'Current version is optimal',\n"," 'suggestion': 'Use force=True to override'\n"," }"],"metadata":{"id":"FxkVT7_IE9fI"},"execution_count":null,"outputs":[]},{"cell_type":"code","execution_count":null,"metadata":{"colab":{"base_uri":"https://localhost:8080/"},"executionInfo":{"elapsed":7989,"status":"ok","timestamp":1754902146917,"user":{"displayName":"Grégoire","userId":"01249298206117462766"},"user_tz":-120},"id":"80mWsWg4Gi7Q","outputId":"a5019cd9-aa9f-4db7-bb62-d1195a16cbd6"},"outputs":[{"output_type":"stream","name":"stdout","text":["\n","============================================================\n","🎯 Enregistrement d'un modèle SPÉCIFIQUE dans le Registry MLflow\n","🆔 Run ID spécifique: b6dd1ecdeb3f4b5c899b332f3c948cd7\n","🔬 Expérience MLflow: OC Projet 8\n","📦 Nom du modèle Registry: StreetsSegmentation\n","⚡ Mode FORCE activé - Mise en prod forcée\n","============================================================\n","🎯 Mode spécifique activé - utilisation du run b6dd1ecdeb3f4b5c899b332f3c948cd7\n","✅ Run spécifique trouvé:\n"," • Run Name: FPN_efficientnetb0_finetune_augTrue_prod\n"," • Run ID: b6dd1ecdeb3f4b5c899b332f3c948cd7\n"," • Test Mean IoU: 0.7457\n"," • Test Accuracy: 0.9123\n","\n","📦 Modèle 'StreetsSegmentation' trouvé dans le Registry\n"," • Version la plus récente: 1\n"," • Stage: Production\n"," • Test Mean IoU: 0.7488\n","\n","🎯 Enregistrement du modèle spécifique...\n"," Enregistrement FORCÉ du modèle spécifique\n"," Enregistrement FORCÉ du modèle spécifique...\n"," 📝 Enregistrement d'une nouvelle version...\n"]},{"output_type":"stream","name":"stderr","text":["2025/08/11 08:49:04 INFO mlflow.store.model_registry.abstract_store: Waiting up to 300 seconds for model version to finish creation. Model name: StreetsSegmentation, version 2\n"]},{"output_type":"stream","name":"stdout","text":[" 📁 Artefact trouvé: best_model.keras\n"," 🏷️ Ajout des tags de métadonnées...\n"," 🚀 Promotion de la version 2 en Production...\n"]},{"output_type":"stream","name":"stderr","text":["/tmp/ipython-input-652925202.py:269: FutureWarning: ``mlflow.tracking.client.MlflowClient.transition_model_version_stage`` is deprecated since 2.9.0. Model registry stages will be removed in a future major release. To learn more about the deprecation of model registry stages, see our migration guide here: https://mlflow.org/docs/latest/model-registry.html#migrating-from-stages\n"," client.transition_model_version_stage(\n"]},{"output_type":"stream","name":"stdout","text":["✅ Modèle enregistré avec succès !\n"," • Nom: StreetsSegmentation\n"," • Version: 2\n"," • Stage: Production\n"," • Test Mean IoU: 0.7457\n"," • Mode de sélection: specific\n"]},{"output_type":"execute_result","data":{"text/plain":["{'model_name': 'StreetsSegmentation',\n"," 'version': '2',\n"," 'run_id': 'b6dd1ecdeb3f4b5c899b332f3c948cd7',\n"," 'test_mean_iou': 0.7457431554794312,\n"," 'selection_mode': 'specific',\n"," 'action': 'updated',\n"," 'tags': {'source_run_id': 'b6dd1ecdeb3f4b5c899b332f3c948cd7',\n"," 'test_mean_iou': '0.7457431554794312',\n"," 'model_name': 'StreetsSegmentation',\n"," 'source_experiment': 'FPN_efficientnetb0_finetune_augTrue_prod',\n"," 'selection_mode': 'specific',\n"," 'registration_mode': 'forced'}}"]},"metadata":{},"execution_count":56}],"source":["# Enregistrement automatique (seulement si meilleur)\n","register_best_model(experiment_name=\"OC Projet 8\", model_name=\"StreetsSegmentation\")"]},{"cell_type":"markdown","metadata":{"id":"m5gJ0G9SK2Pn"},"source":["## Fonction d'inférence et démonstration"]},{"cell_type":"markdown","source":["Les fonctions utiles sont toutes dans les imports en début de notebook"],"metadata":{"id":"IrCVqitDg5sk"}},{"cell_type":"code","execution_count":null,"metadata":{"colab":{"base_uri":"https://localhost:8080/","height":170,"referenced_widgets":["e281656014894d9d9620797934ba5ac7","f471b2bed65b46b7a064af9923caaa9c","5c6880a939f24969b175a6ad56eb3199","ed37df01d0d241ddabd8dda1350cad89","2e9f4d996df84b45bae61884af237962","e5838e9e65dd4986975162d57fb00706","0c9d780c40aa4112b8476a595c8f5ddc","c1bb09b764fb4214a1f6c1f9d05873ec","6a3f0a658b494feabf00895d400d62e7","13a5ab43d16045ca8581f3a71bbce225","743b516205314924b2e4142867f3c816"]},"id":"Az_I9YlrRevb","executionInfo":{"status":"ok","timestamp":1754922747003,"user_tz":-120,"elapsed":21698,"user":{"displayName":"Grégoire","userId":"01249298206117462766"}},"outputId":"87bb4bc5-9d6e-4fa9-f66f-3c20db294b78"},"outputs":[{"output_type":"stream","name":"stdout","text":["📦 Téléchargement du modèle StreetsSegmentation (run: b6dd1ecdeb3f4b5c899b332f3c948cd7)\n"]},{"output_type":"display_data","data":{"text/plain":["Downloading artifacts: 0%| | 0/1 [00:00"],"image/png":"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\n"},"metadata":{}},{"output_type":"stream","name":"stdout","text":["\n","📊 Statistiques du masque prédit:\n"," flat: 6853 pixels (13.7%)\n"," human: 7970 pixels (15.9%)\n"," vehicle: 9227 pixels (18.4%)\n"," construction: 19823 pixels (39.5%)\n"," object: 1727 pixels (3.4%)\n"," nature: 889 pixels (1.8%)\n"," sky: 782 pixels (1.6%)\n"," void: 2905 pixels (5.8%)\n"]}]},{"cell_type":"code","source":["# === EXEMPLE D'UTILISATION ===\n","\n","# 1. Charger la configuration des classes\n","mapping_config = load_cityscapes_config(\"../cityscapes_config.json\")\n","\n","# 2. Faire l'inférence sur une nouvelle image\n","predicted_mask1 = run_inference_and_visualize(\n"," image_path=\"./content/data/test_images_sample/original/frankfurt_000000_000294_leftImg8bit.png\",\n"," mask_path=\"./content/data/test_images_sample/mask/frankfurt_000000_000294_gtFine_labelIds.png\",\n"," model=model,\n"," encoder_name=encoder_name,\n"," img_size=script_img_size,\n"," mapping_config=mapping_config\n",")"],"metadata":{"colab":{"base_uri":"https://localhost:8080/","height":1000},"id":"RqYbA_qD52V2","executionInfo":{"status":"ok","timestamp":1754922759673,"user_tz":-120,"elapsed":4651,"user":{"displayName":"Grégoire","userId":"01249298206117462766"}},"outputId":"212c3cd7-2dad-43b6-901a-829c32f478ef"},"execution_count":null,"outputs":[{"output_type":"stream","name":"stdout","text":["== Groupes de classes Cityscapes ==\n"," - flat : ['road', 'sidewalk', 'parking', 'rail track']\n"," - human : ['person', 'rider']\n"," - vehicle : ['car', 'truck', 'bus', 'train', 'motorcycle', 'bicycle', 'caravan', 'trailer']\n"," - construction : ['building', 'wall', 'fence', 'guard rail', 'bridge', 'tunnel']\n"," - object : ['pole', 'pole group', 'traffic sign', 'traffic light']\n"," - nature : ['vegetation', 'terrain']\n"," - sky : ['sky']\n"," - void : ['unlabeled', 'ego vehicle', 'rectification border', 'out of roi', 'static', 'dynamic', 'ground']\n","\n","== Mapping ID Cityscapes → Groupe ==\n"," ID 0 ('unlabeled') → groupe 'void' (7)\n"," ID 1 ('ego vehicle') → groupe 'void' (7)\n"," ID 2 ('rectification border') → groupe 'void' (7)\n"," ID 3 ('out of roi') → groupe 'void' (7)\n"," ID 4 ('static') → groupe 'void' (7)\n"," ID 5 ('dynamic') → groupe 'void' (7)\n"," ID 6 ('ground') → groupe 'void' (7)\n"," ID 7 ('road') → groupe 'flat' (0)\n"," ID 8 ('sidewalk') → groupe 'flat' (0)\n"," ID 9 ('parking') → groupe 'flat' (0)\n"," ID 10 ('rail track') → groupe 'flat' (0)\n"," ID 11 ('building') → groupe 'construction' (3)\n"," ID 12 ('wall') → groupe 'construction' (3)\n"," ID 13 ('fence') → groupe 'construction' (3)\n"," ID 14 ('guard rail') → groupe 'construction' (3)\n"," ID 15 ('bridge') → groupe 'construction' (3)\n"," ID 16 ('tunnel') → groupe 'construction' (3)\n"," ID 17 ('pole') → groupe 'object' (4)\n"," ID 18 ('pole group') → groupe 'object' (4)\n"," ID 19 ('traffic light') → groupe 'object' (4)\n"," ID 20 ('traffic sign') → groupe 'object' (4)\n"," ID 21 ('vegetation') → groupe 'nature' (5)\n"," ID 22 ('terrain') → groupe 'nature' (5)\n"," ID 23 ('sky') → groupe 'sky' (6)\n"," ID 24 ('person') → groupe 'human' (1)\n"," ID 25 ('rider') → groupe 'human' (1)\n"," ID 26 ('car') → groupe 'vehicle' (2)\n"," ID 27 ('truck') → groupe 'vehicle' (2)\n"," ID 28 ('bus') → groupe 'vehicle' (2)\n"," ID 29 ('caravan') → groupe 'vehicle' (2)\n"," ID 30 ('trailer') → groupe 'vehicle' (2)\n"," ID 31 ('train') → groupe 'vehicle' (2)\n"," ID 32 ('motorcycle') → groupe 'vehicle' (2)\n"," ID 33 ('bicycle') → groupe 'vehicle' (2)\n","\n","== Liste des groupes et couleurs associées ==\n"," Groupe 0: flat - Couleur RGB : [128 64 128]\n"," Groupe 1: human - Couleur RGB : [220 20 60]\n"," Groupe 2: vehicle - Couleur RGB : [ 0 0 142]\n"," Groupe 3: construction - Couleur RGB : [70 70 70]\n"," Groupe 4: object - Couleur RGB : [220 220 0]\n"," Groupe 5: nature - Couleur RGB : [107 142 35]\n"," Groupe 6: sky - Couleur RGB : [ 70 130 180]\n"," Groupe 7: void - Couleur RGB : [0 0 0]\n","\n","Nombre total de groupes/classes : 8\n","=== INFÉRENCE SUR frankfurt_000000_000294_leftImg8bit.png ===\n","🔄 Préprocessing de l'image: frankfurt_000000_000294_leftImg8bit.png\n","ℹ️ Modèle vanilla - pas de preprocessing\n","🔄 Préprocessing du masque: frankfurt_000000_000294_gtFine_labelIds.png\n"," Valeurs uniques avant mapping: [ 3 2 11 23 17 4 20 21 26 24]...\n"," Valeurs uniques après mapping: [7 3 6 4 5 2 1 0]\n","🔮 Exécution de l'inférence...\n","✅ Inférence terminée - Shape: (224, 224)\n"]},{"output_type":"display_data","data":{"text/plain":["\n","\n","- **FPN (Feature Pyramid Network)** 🧭 \n"," FPN est une architecture conçue pour capturer efficacement des objets à **différentes échelles**, grâce à une combinaison intelligente de caractéristiques extraites à plusieurs résolutions.\n","\n"," Elle repose sur deux étapes principales :\n"," \n"," 1. 🔽 **Chemin descendant (bottom-up)** : \n"," L’image passe dans un encodeur profond (souvent un ResNet), qui génère des cartes de caractéristiques à plusieurs résolutions (`1/4`, `1/8`, `1/16`, `1/32`). \n"," Plus la résolution est petite, plus l’information est abstraite.\n"," \n"," 2. 🔼 **Chemin ascendant (top-down)** : \n"," Les cartes profondes sont **upsamplées progressivement** (x2 à chaque étape) et **fusionnées avec des cartes plus superficielles** à l’aide de *connexions latérales* (lateral connections). \n"," Cette fusion permet de combiner des informations **globales et locales** à chaque niveau de la pyramide.\n","\n"," Finalement, les différentes cartes sont combinées et passées dans un petit réseau convolutionnel (souvent un 3x3) pour produire la **prédiction finale**.\n","\n"," 👉 Cette approche rend FPN particulièrement efficace sur des images avec des objets de **tailles très variées**, comme en vision par ordinateur généraliste (urbain, aérien, etc.).\n","\n"," Le modèle a été historiquement entraîné avec ResNet50 et ResNet101 par l'équipe de chercheurs.\n","> **“We build FPN on top of ResNet-50/101 [21], which are used as backbone feature extractors.”** \n","> — *Feature Pyramid Networks for Object Detection*, \n","> Tsung-Yi Lin, Piotr Dollár, Ross Girshick, Kaiming He, Serge Belongie. \n","> *CVPR 2017* \n","> [arXiv:1612.03144](https://arxiv.org/abs/1612.03144)\n","\n","\n"," *(Le schéma ci-dessous illustre cette architecture dans le contexte original de la détection, mais le principe est le même en segmentation sémantique.)*\n","\n"," \n","\n","---\n","\n","### Choix des encodeurs 🧠\n","\n","Plusieurs encodeurs sont disponibles pour ces architectures : \n","- **Encodeur intégré** 🆕 : L’encodeur est entraîné en même temps que le décodeur, sans poids pré-entraînés. Pour U-Net, il a été développé *from scratch* mais pour FPN, il s'agira de ResNet34 (plus léger que ResNet50, le modèle de feature extraction originellement utilisé dans les travaux de recherche).\n","- **MobilenetV2 pré-entraîné** 🚀 : Un encodeur léger et efficace, pré-entraîné sur ImageNet, permettant un bon compromis entre rapidité et performance. \n","- **EfficientNetB0 pré-entraîné** ⚡ : Un encodeur plus puissant et performant, également basé sur un modèle pré-entraîné, utilisé pour tirer parti du transfert learning.\n","\n","ℹ️ Par défaut, la bibliothèque `segmentation_models` utilise **ResNet34 pré-entraîné sur ImageNet** comme encodeur si aucun autre n’est spécifié.\n","\n","---\n","\n","### Gel ou fine-tuning de l’encodeur ❄️🔥\n","\n","Lors de l’entraînement, deux stratégies sont possibles : \n","- **Encodage gelé** ❄️ : Les poids de l’encodeur pré-entraîné restent fixes, ce qui accélère l’entraînement et évite la modification de caractéristiques déjà apprises. \n","- **Encodage entraînable (fine-tuning)** 🔥 : Les poids de l’encodeur sont ajustés pour mieux s’adapter aux données spécifiques, ce qui peut améliorer les performances au prix d’un temps d’entraînement plus long.\n","\n","---\n","\n","### Stratégie recommandée pour l’expérimentation 🎯\n","\n","1. **Tester toutes les architectures et encodeurs avec l’encodeur gelé** ❄️ pour un benchmark rapide. \n","2. **Sélectionner la meilleure configuration** 🏆 selon les performances obtenues. \n","3. **Refaire l’entraînement avec l’encodeur dégelé (fine-tuning)** 🔥 pour optimiser les résultats finaux.\n","\n","Cette méthode permet de gagner du temps tout en explorant efficacement l’espace des modèles.\n","\n"]},{"cell_type":"markdown","metadata":{"id":"s5te0Fcwtx51"},"source":["### Création des modèles"]},{"cell_type":"code","execution_count":15,"metadata":{"colab":{"base_uri":"https://localhost:8080/"},"executionInfo":{"elapsed":5,"status":"ok","timestamp":1755000398851,"user":{"displayName":"Grégoire Mureau","userId":"04855910922109670788"},"user_tz":-120},"id":"8eHYsAmIRsIU","outputId":"786a2465-6e3a-4278-e5da-f46912fac4f9"},"outputs":[{"output_type":"stream","name":"stdout","text":["🔧 BACKBONES DISPONIBLES:\n","==================================================\n","⭐ RECOMMANDÉS pour voiture autonome (vitesse/précision):\n"," • mobilenetv2\n"," • efficientnetb0\n"," • resnet34\n"," • resnet50\n","\n","📋 TOUS DISPONIBLES (32):\n","\n"," resnet18 resnet34 resnet50 resnet101 \n"," resnet152 seresnet18 seresnet34 seresnet50 \n"," seresnet101 seresnet152 seresnext50 seresnext101 \n"," senet154 resnext50 resnext101 vgg16 \n"," vgg19 densenet121 densenet169 densenet201 \n"," inceptionresnetv2 inceptionv3 mobilenet mobilenetv2 \n"," efficientnetb0 efficientnetb1 efficientnetb2 efficientnetb3 \n"," efficientnetb4 efficientnetb5 efficientnetb6 efficientnetb7 \n"]}],"source":["def show_available_backbones():\n"," \"\"\"Affiche tous les backbones disponibles dans segmentation_models\"\"\"\n","\n"," # Essayer de récupérer la liste des backbones disponibles\n"," try:\n"," available_backbones = sm.get_available_backbones()\n","\n"," print(\"🔧 BACKBONES DISPONIBLES DANS SEGMENTATION_MODELS:\")\n"," print(\"=\" * 60)\n","\n"," # Grouper par famille\n"," families = {\n"," 'ResNet': [b for b in available_backbones if b.startswith('resnet')],\n"," 'SEResNet': [b for b in available_backbones if b.startswith('seresnet')],\n"," 'SEResNeXt': [b for b in available_backbones if b.startswith('seresnext')],\n"," 'ResNeXt': [b for b in available_backbones if b.startswith('resnext')],\n"," 'VGG': [b for b in available_backbones if b.startswith('vgg')],\n"," 'DenseNet': [b for b in available_backbones if b.startswith('densenet')],\n"," 'Inception': [b for b in available_backbones if 'inception' in b],\n"," 'MobileNet': [b for b in available_backbones if 'mobilenet' in b],\n"," 'EfficientNet': [b for b in available_backbones if 'efficientnet' in b],\n"," 'SENet': [b for b in available_backbones if b.startswith('senet')],\n"," }\n","\n"," for family, models in families.items():\n"," if models:\n"," print(f\"\\n📱 {family}:\")\n"," for model in sorted(models):\n"," print(f\" • {model}\")\n","\n"," except AttributeError:\n"," # Si la méthode n'existe pas, utiliser la liste connue\n"," known_backbones = [\n"," 'resnet18', 'resnet34', 'resnet50', 'resnet101', 'resnet152',\n"," 'seresnet18', 'seresnet34', 'seresnet50', 'seresnet101', 'seresnet152',\n"," 'seresnext50', 'seresnext101', 'senet154', 'resnext50', 'resnext101',\n"," 'vgg16', 'vgg19', 'densenet121', 'densenet169', 'densenet201',\n"," 'inceptionresnetv2', 'inceptionv3', 'mobilenet', 'mobilenetv2',\n"," 'efficientnetb0', 'efficientnetb1', 'efficientnetb2', 'efficientnetb3',\n"," 'efficientnetb4', 'efficientnetb5', 'efficientnetb6', 'efficientnetb7'\n"," ]\n","\n"," print(\"🔧 BACKBONES DISPONIBLES:\")\n"," print(\"=\" * 50)\n","\n"," # Recommandations pour voiture autonome\n"," recommended = ['mobilenetv2', 'efficientnetb0', 'resnet34', 'resnet50']\n","\n"," print(\"⭐ RECOMMANDÉS pour voiture autonome (vitesse/précision):\")\n"," for model in recommended:\n"," print(f\" • {model}\")\n","\n"," print(f\"\\n📋 TOUS DISPONIBLES ({len(known_backbones)}):\")\n"," for i, model in enumerate(known_backbones):\n"," if i % 4 == 0: # Nouvelle ligne tous les 4\n"," print()\n"," print(f\" {model:<20}\", end=\"\")\n"," print() # Ligne finale\n","\n","# Afficher les backbones disponibles\n","show_available_backbones()"]},{"cell_type":"code","execution_count":22,"metadata":{"id":"3xLKhC8KW5i7","executionInfo":{"status":"ok","timestamp":1755008390123,"user_tz":-120,"elapsed":73,"user":{"displayName":"Grégoire Mureau","userId":"04855910922109670788"}}},"outputs":[],"source":["# ================================================================\n","# GRILLE D'EXPÉRIMENTATIONS - SEGMENTATION SÉMANTIQUE CITYSCAPES\n","# ================================================================\n","\n","# Architectures et backbones à tester\n","model_variants = [\n"," # Baselines (sans backbone pré-entraîné)\n"," ('U-Net', None),\n"," ('FPN', None), # --> l'encodeur choisi sera alors vgg16\n"," # Modèles avec backbones pré-entraînés (encoders gelés)\n"," ('U-Net', 'mobilenetv2'),\n"," ('U-Net', 'efficientnetb0'),\n"," ('FPN', 'mobilenetv2'),\n"," ('FPN', 'efficientnetb0'),\n"," ('FPN', 'resnet34'),\n","]\n","\n","# Fine-tuning sélectif (encoders entraînables)\n","finetune_variants = [\n"," ('U-Net', 'mobilenetv2'),\n"," ('FPN', 'efficientnetb0'),\n"," ('FPN', 'resnet34'),\n","]\n","\n","# Paramètres d'entraînement\n","base_params = {\n"," 'learning_rate': 1e-4,\n"," 'epochs': 20,\n"," 'batch_size': 8,\n"," 'img_size': (224, 224)\n","}\n","\n","finetune_params = {\n"," 'learning_rate': 5e-5, # Plus petit pour le fine-tuning\n"," 'epochs': 30,\n"," 'batch_size': 8,\n"," 'img_size': (224, 224)\n","}\n","\n","# Génération des configurations\n","def generate_experiment_configs():\n"," \"\"\"Génère toutes les configurations d'expérimentations\"\"\"\n"," model_configs = []\n","\n"," # 1. Expériences de base (encodeurs gelés ou vanilla)\n"," for arch, encoder in model_variants:\n"," for augment in [False, True]:\n"," # ✅ Nom correct selon le type\n"," if encoder is None:\n"," config_name = f\"{arch}_vanilla_aug{augment}\"\n"," else:\n"," config_name = f\"{arch}_{encoder}_frozen_aug{augment}\"\n","\n"," model_configs.append({\n"," \"name\": config_name,\n"," \"architecture\": arch,\n"," \"encoder\": encoder,\n"," \"encoder_freeze\": encoder is not None,\n"," \"augmentation\": augment,\n"," \"params\": base_params.copy(),\n"," \"type\": \"baseline\" if encoder is None else \"frozen_pretrained\"\n"," })\n","\n"," # 2. Expériences de fine-tuning (encoders dégelés)\n"," for arch, encoder in finetune_variants:\n"," config_name = f\"{arch}_{encoder}_finetune_augTrue\"\n"," model_configs.append({\n"," \"name\": config_name,\n"," \"architecture\": arch,\n"," \"encoder\": encoder,\n"," \"encoder_freeze\": False, # False = dégelé pour fine-tuning\n"," \"augmentation\": True,\n"," \"params\": finetune_params.copy(),\n"," \"type\": \"finetuned_pretrained\"\n"," })\n","\n"," return model_configs"]},{"cell_type":"code","execution_count":23,"metadata":{"id":"X7wY6OqSZFha","executionInfo":{"status":"ok","timestamp":1755008390314,"user_tz":-120,"elapsed":56,"user":{"displayName":"Grégoire Mureau","userId":"04855910922109670788"}}},"outputs":[],"source":["def create_model_from_config(config, num_classes, verbose=False):\n"," \"\"\"\n"," Crée un modèle de segmentation à partir d'une configuration d'expérimentation.\n"," \"\"\"\n"," # === EXTRACTION DES PARAMÈTRES ===\n"," architecture = config['architecture']\n"," encoder = config['encoder']\n"," encoder_freeze = config['encoder_freeze']\n"," learning_rate = config['params']['learning_rate']\n"," img_size = config['params']['img_size']\n","\n"," # === MAPPING DES ARCHITECTURES ===\n"," model_classes = {\n"," 'U-Net': sm.Unet,\n"," 'FPN': sm.FPN,\n"," }\n","\n"," if architecture not in model_classes:\n"," raise ValueError(f\"Architecture '{architecture}' non supportée.\")\n","\n"," ModelClass = model_classes[architecture]\n"," input_shape = (*img_size, 3)\n","\n"," if verbose:\n"," print(f\"🔧 Création du modèle '{config['name']}'\")\n"," print(f\" Architecture: {architecture}\")\n"," print(f\" Encoder: {encoder or 'Aucun (vanilla)'}\")\n"," print(f\" Encoder gelé: {encoder_freeze}\")\n","\n"," try:\n"," if encoder is None:\n"," # === MODÈLE VANILLA ===\n"," model = ModelClass(\n"," classes=num_classes,\n"," activation='softmax',\n"," input_shape=input_shape,\n"," )\n"," else:\n"," # === MODÈLE AVEC BACKBONE ===\n"," model = ModelClass(\n"," backbone_name=encoder,\n"," encoder_weights='imagenet',\n"," encoder_freeze=encoder_freeze,\n"," classes=num_classes,\n"," activation='softmax',\n"," input_shape=input_shape,\n"," )\n","\n"," # === COMPILATION ===\n"," model.compile(\n"," optimizer=keras.optimizers.Adam(learning_rate=learning_rate),\n"," loss=sm.losses.categorical_focal_jaccard_loss,\n"," metrics=[sm.metrics.iou_score, 'accuracy']\n"," )\n","\n"," if verbose:\n"," total_params = model.count_params()\n"," trainable_params = sum([tf.keras.backend.count_params(w) for w in model.trainable_weights])\n"," frozen_params = total_params - trainable_params\n"," print(f\" 📊 Paramètres: {total_params:,} total | {trainable_params:,} entraînables | {frozen_params:,} gelés\")\n","\n"," return model\n","\n"," except Exception as e:\n"," print(f\"❌ Erreur: {e}\")\n"," raise"]},{"cell_type":"code","execution_count":24,"metadata":{"id":"4ROnd_8f9_W7","executionInfo":{"status":"ok","timestamp":1755008390430,"user_tz":-120,"elapsed":14,"user":{"displayName":"Grégoire Mureau","userId":"04855910922109670788"}}},"outputs":[],"source":["# Génération des configurations\n","experimental_configurations = generate_experiment_configs()"]},{"cell_type":"markdown","metadata":{"id":"GILXYzC_9jHX"},"source":["### Analyse et compréhension détaillée des modèles"]},{"cell_type":"markdown","metadata":{"id":"5yr47E0qbmmY"},"source":["L'objectif est simplement ici d'afficher des informations générales sur les modèles et notamment vérifier le gel des poids pour les modèles de feature extraction (backbone) qui fournissent des descripteurs aux modèles de segmentation et ce afin de pouvoir comparer les différences de performances."]},{"cell_type":"code","execution_count":19,"metadata":{"id":"qdsGkrVQe2mX","executionInfo":{"status":"ok","timestamp":1755000398995,"user_tz":-120,"elapsed":47,"user":{"displayName":"Grégoire Mureau","userId":"04855910922109670788"}}},"outputs":[],"source":["def analyze_model_comprehensive(config, num_classes, show_architecture=False, detailed_layers=False):\n"," \"\"\"\n"," Analyse exhaustive d'un modèle de segmentation avec affichage clair des couches gelées/dégelées\n","\n"," Args:\n"," config (dict): Configuration du modèle\n"," num_classes (int): Nombre de classes\n"," show_architecture (bool): Afficher model.summary()\n"," detailed_layers (bool): Afficher TOUTES les couches (sinon juste backbone + résumé)\n","\n"," Returns:\n"," dict: Résultats de l'analyse\n"," \"\"\"\n"," print(f\"\\n{'='*70}\")\n"," print(f\"🏗️ ANALYSE COMPLÈTE: {config['name']}\")\n"," print(f\"{'='*70}\")\n","\n"," # === 1. INFORMATIONS DE CONFIGURATION ===\n"," print(f\"📊 Configuration:\")\n"," print(f\" • Architecture: {config['architecture']}\")\n"," print(f\" • Backbone: {config['encoder'] or 'Aucun (vanilla)'}\")\n"," print(f\" • Encoder gelé: {'✅ Oui' if config.get('encoder_freeze', False) else '❌ Non'}\")\n"," print(f\" • Type: {config['type'].replace('_', ' ').title()}\")\n"," print(f\" • Augmentation: {'✅ Oui' if config['augmentation'] else '❌ Non'}\")\n"," print(f\" • Learning rate: {config['params']['learning_rate']}\")\n"," print(f\" • Époques prévues: {config['params']['epochs']}\")\n"," print(f\" • Taille images: {config['params']['img_size']}\")\n","\n"," try:\n"," # === 2. CRÉATION DU MODÈLE ===\n"," print(f\"\\n🔧 Création du modèle...\")\n"," model = create_model_from_config(config, num_classes, verbose=False)\n","\n"," # === 3. ANALYSE DES PARAMÈTRES ===\n"," total_params = model.count_params()\n"," trainable_params = sum([tf.keras.backend.count_params(w) for w in model.trainable_weights])\n"," frozen_params = total_params - trainable_params\n"," frozen_ratio = (frozen_params / total_params) * 100\n","\n"," print(f\"\\n🔢 Paramètres:\")\n"," print(f\" • Total: {total_params:,}\")\n"," print(f\" • Entraînables: {trainable_params:,} ({(trainable_params/total_params)*100:.1f}%)\")\n"," print(f\" • Gelés: {frozen_params:,} ({frozen_ratio:.1f}%)\")\n","\n"," # === 4. ANALYSE DES COUCHES ===\n"," backbone_layers = []\n"," other_layers = []\n"," backbone_total = 0\n"," backbone_frozen = 0\n","\n"," # Classifier les couches\n"," for layer in model.layers:\n"," layer_params = layer.count_params()\n"," is_backbone = False\n","\n"," if config['encoder']:\n"," backbone_keywords = [\n"," config['encoder'].lower(),\n"," 'encoder',\n"," 'backbone',\n"," 'feature'\n"," ]\n"," is_backbone = any(keyword in layer.name.lower() for keyword in backbone_keywords)\n","\n"," layer_info = {\n"," 'name': layer.name,\n"," 'params': layer_params,\n"," 'trainable': layer.trainable,\n"," 'is_backbone': is_backbone\n"," }\n","\n"," if is_backbone:\n"," backbone_layers.append(layer_info)\n"," backbone_total += layer_params\n"," if not layer.trainable:\n"," backbone_frozen += layer_params\n"," else:\n"," other_layers.append(layer_info)\n","\n"," # === 5. AFFICHAGE DES COUCHES ===\n"," if detailed_layers:\n"," # Affichage complet de toutes les couches\n"," print(f\"\\n🔍 TOUTES LES COUCHES (🟢 = Entraînable, 🔴 = Gelé):\")\n"," print(\"-\" * 70)\n","\n"," all_layers = backbone_layers + other_layers\n"," for i, layer_info in enumerate(all_layers):\n"," icon = \"🔴\" if not layer_info['trainable'] else \"🟢\"\n"," status = \"GELÉ\" if not layer_info['trainable'] else \"ENTRAÎNABLE\"\n","\n"," # Identifier le type de couche\n"," if layer_info['is_backbone']:\n"," layer_type = \" (BACKBONE)\"\n"," elif 'decoder' in layer_info['name'].lower():\n"," layer_type = \" (DECODER)\"\n"," elif 'classifier' in layer_info['name'].lower() or 'final' in layer_info['name'].lower():\n"," layer_type = \" (CLASSIFICATION)\"\n"," elif 'upsample' in layer_info['name'].lower():\n"," layer_type = \" (UPSAMPLING)\"\n"," else:\n"," layer_type = \"\"\n","\n"," print(f\"{i+1:2d}. {icon} {layer_info['name']:<40} {layer_info['params']:>8,} params - {status}{layer_type}\")\n","\n"," else:\n"," # Affichage résumé avec focus sur le backbone\n"," if backbone_layers:\n"," print(f\"\\n🔍 COUCHES DU BACKBONE (🟢 = Entraînable, 🔴 = Gelé):\")\n"," print(\"-\" * 60)\n","\n"," for layer_info in backbone_layers[:8]: # Limite à 8 pour lisibilité\n"," icon = \"🔴\" if not layer_info['trainable'] else \"🟢\"\n"," status = \"GELÉ\" if not layer_info['trainable'] else \"ENTRAÎNABLE\"\n"," print(f\" {icon} {layer_info['name']:<35} {layer_info['params']:>8,} params - {status}\")\n","\n"," if len(backbone_layers) > 8:\n"," print(f\" ... et {len(backbone_layers)-8} autres couches du backbone\")\n","\n"," # Résumé des autres couches\n"," other_frozen = sum(1 for layer in other_layers if not layer['trainable'])\n"," other_trainable = len(other_layers) - other_frozen\n"," print(f\"\\n🏗️ AUTRES COUCHES (résumé):\")\n"," print(f\" • {other_trainable} couches entraînables\")\n"," print(f\" • {other_frozen} couches gelées\")\n"," print(f\" • (Utilisez detailed_layers=True pour voir toutes les couches)\")\n","\n"," # === 6. RÉSUMÉ BACKBONE ===\n"," if backbone_layers:\n"," backbone_frozen_ratio = (backbone_frozen / backbone_total * 100) if backbone_total > 0 else 0\n"," print(f\"\\n📋 Résumé backbone:\")\n"," print(f\" • Total backbone: {backbone_total:,} params\")\n"," print(f\" • Backbone gelé: {backbone_frozen:,} params ({backbone_frozen_ratio:.1f}%)\")\n"," print(f\" • Backbone entraînable: {backbone_total - backbone_frozen:,} params ({100-backbone_frozen_ratio:.1f}%)\")\n","\n"," # === 7. VALIDATION DU GEL ===\n"," if config['encoder']:\n"," print(f\"\\n🎯 Validation du gel du backbone:\")\n","\n"," # Seuils adaptatifs selon le backbone\n"," backbone_name = config['encoder'].lower()\n"," if 'mobilenet' in backbone_name:\n"," expected_frozen_range = (20, 35)\n"," elif 'efficientnet' in backbone_name and 'b0' in backbone_name:\n"," expected_frozen_range = (15, 25)\n"," elif 'resnet' in backbone_name:\n"," expected_frozen_range = (18, 28)\n"," else:\n"," expected_frozen_range = (10, 30)\n","\n"," if config.get('encoder_freeze', False):\n"," # Modèle supposé gelé\n"," if frozen_ratio >= expected_frozen_range[0]:\n"," status = \"✅ OK\"\n"," message = f\"Backbone correctement gelé ({frozen_ratio:.1f}% ≥ {expected_frozen_range[0]}%)\"\n"," else:\n"," status = \"⚠️ SUSPECT\"\n"," message = f\"Gel potentiellement incomplet ({frozen_ratio:.1f}% < {expected_frozen_range[0]}%)\"\n"," else:\n"," # Modèle supposé dégelé\n"," if frozen_ratio < 5.0:\n"," status = \"✅ OK\"\n"," message = f\"Backbone correctement dégelé ({frozen_ratio:.1f}% < 5%)\"\n"," else:\n"," status = \"⚠️ SUSPECT\"\n"," message = f\"Gel résiduel inattendu ({frozen_ratio:.1f}% ≥ 5%)\"\n","\n"," print(f\" • Statut global: {status}\")\n"," print(f\" • {message}\")\n"," print(f\" • Plage attendue pour {config['encoder']}: {expected_frozen_range[0]}-{expected_frozen_range[1]}%\")\n","\n"," # Validation spécifique du backbone\n"," if backbone_total > 0:\n"," backbone_expected_frozen = config.get('encoder_freeze', False)\n"," backbone_actually_frozen = (backbone_frozen / backbone_total) > 0.9\n","\n"," if backbone_expected_frozen == backbone_actually_frozen:\n"," backbone_status = \"✅ COHÉRENT\"\n"," else:\n"," backbone_status = \"❌ INCOHÉRENT\"\n","\n"," print(f\" • Cohérence backbone: {backbone_status}\")\n"," print(f\" - Configuration: {'gelé' if backbone_expected_frozen else 'dégelé'}\")\n"," print(f\" - Réalité: {'gelé' if backbone_actually_frozen else 'dégelé'} ({backbone_frozen_ratio:.1f}%)\")\n","\n"," # === 8. ARCHITECTURE DÉTAILLÉE (optionnel) ===\n"," if show_architecture:\n"," print(f\"\\n🏛️ Architecture détaillée (model.summary()):\")\n"," print(\"-\" * 60)\n"," model.summary(line_length=80)\n","\n"," # === 9. RÉSULTATS POUR RETOUR ===\n"," results = {\n"," 'model': model,\n"," 'total_params': total_params,\n"," 'trainable_params': trainable_params,\n"," 'frozen_params': frozen_params,\n"," 'frozen_ratio': frozen_ratio,\n"," 'backbone_params': backbone_total,\n"," 'backbone_frozen': backbone_frozen,\n"," 'backbone_frozen_ratio': backbone_frozen_ratio if backbone_total > 0 else 0,\n"," 'status': 'success',\n"," 'validation': status if config['encoder'] else 'N/A'\n"," }\n","\n"," print(f\"\\n✅ Analyse terminée avec succès\")\n"," return results\n","\n"," except Exception as e:\n"," print(f\"\\n❌ Erreur lors de l'analyse: {e}\")\n"," return {\n"," 'model': None,\n"," 'status': 'error',\n"," 'error': str(e)\n"," }\n","\n"," finally:\n"," # Nettoyage mémoire\n"," if 'model' in locals():\n"," del model\n"," tf.keras.backend.clear_session()\n","\n","# === FONCTION SÉPARÉE POUR VOIR UNIQUEMENT LES COUCHES (différente maintenant) ===\n","def show_all_layers_status(config, num_classes):\n"," \"\"\"\n"," Fonction spécialisée pour afficher UNIQUEMENT les couches, de manière compacte\n"," \"\"\"\n"," print(f\"\\n🔍 COUCHES - {config['name']}\")\n"," print(\"=\" * 70)\n"," print(\"🟢 ENTRAÎNABLE | 🔴 GELÉ\")\n"," print(\"-\" * 70)\n","\n"," try:\n"," model = create_model_from_config(config, num_classes, verbose=False)\n","\n"," trainable_count = 0\n"," frozen_count = 0\n","\n"," for i, layer in enumerate(model.layers):\n"," layer_params = layer.count_params()\n"," icon = \"🔴\" if not layer.trainable else \"🟢\"\n","\n"," if layer.trainable:\n"," trainable_count += 1\n"," else:\n"," frozen_count += 1\n","\n"," # Type de couche (plus concis)\n"," layer_type = \"\"\n"," if config['encoder'] and any(keyword in layer.name.lower()\n"," for keyword in [config['encoder'].lower(), 'encoder', 'backbone']):\n"," layer_type = \"BACKBONE\"\n"," elif 'decoder' in layer.name.lower():\n"," layer_type = \"DECODER\"\n"," elif 'classifier' in layer.name.lower():\n"," layer_type = \"CLASSIF\"\n"," else:\n"," layer_type = \"AUTRE\"\n","\n"," print(f\"{i+1:2d}. {icon} {layer.name:<35} {layer_params:>8,} {layer_type}\")\n","\n"," print(\"-\" * 70)\n"," print(f\"📊 RÉSUMÉ: {trainable_count} entraînables, {frozen_count} gelés\")\n","\n"," except Exception as e:\n"," print(f\"❌ Erreur: {e}\")\n"," finally:\n"," if 'model' in locals():\n"," del model\n"," tf.keras.backend.clear_session()"]},{"cell_type":"markdown","metadata":{"id":"RYR1STLzZQzS"},"source":["L'objectif est simplement ici de tester le gel des poids pour les modèles de feature extraction (backbone) qui fournissent des descripteurs aux modèles de segmentation et ce afin de pouvoir comparer les différences de performances."]},{"cell_type":"code","execution_count":20,"metadata":{"id":"Jyxe-hvRZJ5C","executionInfo":{"status":"ok","timestamp":1755000399016,"user_tz":-120,"elapsed":20,"user":{"displayName":"Grégoire Mureau","userId":"04855910922109670788"}}},"outputs":[],"source":["def test_backbone_freezing(num_classes=19):\n"," \"\"\"\n"," Fonction de test pour vérifier que le gel du backbone fonctionne correctement.\n"," Version corrigée utilisant encoder_freeze.\n"," \"\"\"\n"," print(\"🧪 TEST DU GEL DU BACKBONE (VERSION CORRIGÉE)\")\n"," print(\"=\" * 60)\n","\n"," # === CONFIGURATIONS DE TEST ===\n"," test_configs = [\n"," {\n"," 'name': 'U-Net_mobilenetv2_frozen_test',\n"," 'architecture': 'U-Net',\n"," 'encoder': 'mobilenetv2',\n"," 'encoder_freeze': True, # GELÉ - devrait avoir beaucoup de params gelés\n"," 'params': {'learning_rate': 1e-4, 'img_size': (224, 224)},\n"," 'type': 'frozen_test'\n"," },\n"," {\n"," 'name': 'U-Net_mobilenetv2_trainable_test',\n"," 'architecture': 'U-Net',\n"," 'encoder': 'mobilenetv2',\n"," 'encoder_freeze': False, # DÉGELÉ - devrait avoir peu de params gelés\n"," 'params': {'learning_rate': 1e-4, 'img_size': (224, 224)},\n"," 'type': 'trainable_test'\n"," }\n"," ]\n","\n"," results = {}\n","\n"," # === EXÉCUTION DES TESTS ===\n"," for config in test_configs:\n"," print(f\"\\n🔍 Test: {config['name']}\")\n"," print(\"-\" * 50)\n","\n"," # Création du modèle avec verbose pour voir les détails\n"," model = create_model_from_config(config, num_classes, verbose=True)\n","\n"," # === COMPTAGE DÉTAILLÉ DES PARAMÈTRES ===\n"," total_params = model.count_params()\n"," trainable_params = sum([tf.keras.backend.count_params(w) for w in model.trainable_weights])\n"," frozen_params = total_params - trainable_params\n"," frozen_ratio = (frozen_params / total_params) * 100\n","\n"," print(f\"\\n📈 Résultat:\")\n"," print(f\" • Total: {total_params:,}\")\n"," print(f\" • Entraînables: {trainable_params:,}\")\n"," print(f\" • Gelés: {frozen_params:,}\")\n"," print(f\" • % gelés: {frozen_ratio:.1f}%\")\n","\n"," # === VALIDATION DU RÉSULTAT ===\n"," if config['encoder_freeze']:\n"," # Pour un modèle gelé, on s'attend à >15% de paramètres gelés\n"," expected_frozen_ratio = 20.0\n"," status = \"✅ OK\" if frozen_ratio > expected_frozen_ratio else \"❌ ÉCHEC\"\n"," print(f\" • Statut (gelé attendu >20%): {status}\")\n"," else:\n"," # Pour un modèle dégelé, on s'attend à <10% de paramètres gelés\n"," expected_frozen_ratio = 5\n"," status = \"✅ OK\" if frozen_ratio < expected_frozen_ratio else \"❌ ÉCHEC\"\n"," print(f\" • Statut (dégelé attendu <5%): {status}\")\n","\n"," # Stockage des résultats pour comparaison\n"," results[config['type']] = {\n"," 'total': total_params,\n"," 'trainable': trainable_params,\n"," 'frozen': frozen_params,\n"," 'frozen_ratio': frozen_ratio\n"," }\n","\n"," # === NETTOYAGE MÉMOIRE ===\n"," del model\n"," tf.keras.backend.clear_session()\n","\n"," # === RÉSUMÉ COMPARATIF ===\n"," print(f\"\\n🎯 RÉSUMÉ COMPARATIF\")\n"," print(\"=\" * 60)\n","\n"," if 'frozen_test' in results and 'trainable_test' in results:\n"," frozen_result = results['frozen_test']\n"," trainable_result = results['trainable_test']\n","\n"," print(f\"Modèle gelé (encoder_freeze=True):\")\n"," print(f\" • Params gelés: {frozen_result['frozen']:,} ({frozen_result['frozen_ratio']:.1f}%)\")\n","\n"," print(f\"Modèle dégelé (encoder_freeze=False):\")\n"," print(f\" • Params gelés: {trainable_result['frozen']:,} ({trainable_result['frozen_ratio']:.1f}%)\")\n","\n"," difference = frozen_result['frozen'] - trainable_result['frozen']\n"," print(f\"Différence: {difference:,} paramètres correctement gelés\")\n","\n"," if difference > 100000: # Au moins 100k params de différence\n"," print(\"✅ TEST GLOBAL: Le gel du backbone fonctionne correctement!\")\n"," else:\n"," print(\"❌ TEST GLOBAL: Le gel du backbone ne semble pas fonctionner!\")"]},{"cell_type":"markdown","metadata":{"id":"EvurXI0XZzng"},"source":["On va lister les différentes expérimentations et vérifier le nombre de poids gelés pour deux configurations différentes : une avec backbone entrainable, l'autre non."]},{"cell_type":"code","execution_count":21,"metadata":{"colab":{"base_uri":"https://localhost:8080/"},"executionInfo":{"elapsed":5094,"status":"ok","timestamp":1755000404122,"user":{"displayName":"Grégoire Mureau","userId":"04855910922109670788"},"user_tz":-120},"id":"AsgetKA1ZM0y","outputId":"e141d21c-f3ff-4a3a-ade8-3aa148fba207"},"outputs":[{"output_type":"stream","name":"stdout","text":["📊 PLAN D'EXPÉRIMENTATIONS - 17 configurations\n","======================================================================\n","🔧 Backbones utilisés: ['efficientnetb0', 'mobilenetv2', 'resnet34']\n","🏗️ Architectures: U-Net, FPN\n","\n","🔹 BASELINE (4 configs):\n"," • U-Net_vanilla_augFalse \n"," • U-Net_vanilla_augTrue \n"," • FPN_vanilla_augFalse \n"," • FPN_vanilla_augTrue \n","\n","🔹 FROZEN PRETRAINED (10 configs):\n"," • U-Net_mobilenetv2_frozen_augFalse (gelé)\n"," • U-Net_mobilenetv2_frozen_augTrue (gelé)\n"," • U-Net_efficientnetb0_frozen_augFalse (gelé)\n"," • U-Net_efficientnetb0_frozen_augTrue (gelé)\n"," • FPN_mobilenetv2_frozen_augFalse (gelé)\n"," • FPN_mobilenetv2_frozen_augTrue (gelé)\n"," • FPN_efficientnetb0_frozen_augFalse (gelé)\n"," • FPN_efficientnetb0_frozen_augTrue (gelé)\n"," • FPN_resnet34_frozen_augFalse (gelé)\n"," • FPN_resnet34_frozen_augTrue (gelé)\n","\n","🔹 FINETUNED PRETRAINED (3 configs):\n"," • U-Net_mobilenetv2_finetune_augTrue (dégelé)\n"," • FPN_efficientnetb0_finetune_augTrue (dégelé)\n"," • FPN_resnet34_finetune_augTrue (dégelé)\n","\n","======================================================================\n","LANCEMENT DU TEST DE VALIDATION\n","======================================================================\n","🧪 TEST DU GEL DU BACKBONE (VERSION CORRIGÉE)\n","============================================================\n","\n","🔍 Test: U-Net_mobilenetv2_frozen_test\n","--------------------------------------------------\n","🔧 Création du modèle 'U-Net_mobilenetv2_frozen_test'\n"," Architecture: U-Net\n"," Encoder: mobilenetv2\n"," Encoder gelé: True\n","Downloading data from https://github.com/JonathanCMitchell/mobilenet_v2_keras/releases/download/v1.1/mobilenet_v2_weights_tf_dim_ordering_tf_kernels_1.0_224_no_top.h5\n","\u001b[1m9406464/9406464\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 0us/step\n"," 📊 Paramètres: 8,050,051 total | 5,824,195 entraînables | 2,225,856 gelés\n","\n","📈 Résultat:\n"," • Total: 8,050,051\n"," • Entraînables: 5,824,195\n"," • Gelés: 2,225,856\n"," • % gelés: 27.7%\n"," • Statut (gelé attendu >20%): ✅ OK\n","\n","🔍 Test: U-Net_mobilenetv2_trainable_test\n","--------------------------------------------------\n","🔧 Création du modèle 'U-Net_mobilenetv2_trainable_test'\n"," Architecture: U-Net\n"," Encoder: mobilenetv2\n"," Encoder gelé: False\n"," 📊 Paramètres: 8,050,051 total | 8,013,955 entraînables | 36,096 gelés\n","\n","📈 Résultat:\n"," • Total: 8,050,051\n"," • Entraînables: 8,013,955\n"," • Gelés: 36,096\n"," • % gelés: 0.4%\n"," • Statut (dégelé attendu <5%): ✅ OK\n","\n","🎯 RÉSUMÉ COMPARATIF\n","============================================================\n","Modèle gelé (encoder_freeze=True):\n"," • Params gelés: 2,225,856 (27.7%)\n","Modèle dégelé (encoder_freeze=False):\n"," • Params gelés: 36,096 (0.4%)\n","Différence: 2,189,760 paramètres correctement gelés\n","✅ TEST GLOBAL: Le gel du backbone fonctionne correctement!\n"]}],"source":["# Affichage des configurations\n","print(f\"📊 PLAN D'EXPÉRIMENTATIONS - {len(experimental_configurations)} configurations\")\n","print(\"=\" * 70)\n","\n","# Affichage des backbones utilisés\n","used_backbones = set()\n","for config in experimental_configurations:\n"," if config['encoder']:\n"," used_backbones.add(config['encoder'])\n","\n","print(f\"🔧 Backbones utilisés: {sorted(used_backbones)}\")\n","print(f\"🏗️ Architectures: U-Net, FPN\")\n","\n","# Grouper par type pour un affichage plus clair\n","config_types = {}\n","for config in experimental_configurations:\n"," config_type = config['type']\n"," if config_type not in config_types:\n"," config_types[config_type] = []\n"," config_types[config_type].append(config)\n","\n","for type_name, configs in config_types.items():\n"," print(f\"\\n🔹 {type_name.upper().replace('_', ' ')} ({len(configs)} configs):\")\n"," for config in configs:\n"," freeze_status = \"gelé\" if config['encoder_freeze'] else \"dégelé\"\n"," encoder_info = f\"({freeze_status})\" if config['encoder'] else \"\"\n"," print(f\" • {config['name']} {encoder_info}\")\n","\n","# === LANCEMENT DU TEST ===\n","print(f\"\\n\" + \"=\"*70)\n","print(\"LANCEMENT DU TEST DE VALIDATION\")\n","print(\"=\"*70)\n","\n","# Décommentez pour exécuter le test\n","test_backbone_freezing(num_classes=19)"]},{"cell_type":"code","execution_count":22,"metadata":{"colab":{"base_uri":"https://localhost:8080/"},"executionInfo":{"elapsed":4575,"status":"ok","timestamp":1755000408705,"user":{"displayName":"Grégoire Mureau","userId":"04855910922109670788"},"user_tz":-120},"id":"9TWHaWHA6z66","outputId":"76ed8e6a-552e-4abe-db2b-e98d71569425"},"outputs":[{"output_type":"stream","name":"stdout","text":["\n","======================================================================\n","🏗️ ANALYSE COMPLÈTE: U-Net_mobilenetv2_frozen_augFalse\n","======================================================================\n","📊 Configuration:\n"," • Architecture: U-Net\n"," • Backbone: mobilenetv2\n"," • Encoder gelé: ✅ Oui\n"," • Type: Frozen Pretrained\n"," • Augmentation: ❌ Non\n"," • Learning rate: 0.0001\n"," • Époques prévues: 20\n"," • Taille images: (224, 224)\n","\n","🔧 Création du modèle...\n","\n","🔢 Paramètres:\n"," • Total: 8,050,051\n"," • Entraînables: 5,824,195 (72.3%)\n"," • Gelés: 2,225,856 (27.7%)\n","\n","🏗️ AUTRES COUCHES (résumé):\n"," • 93 couches entraînables\n"," • 103 couches gelées\n"," • (Utilisez detailed_layers=True pour voir toutes les couches)\n","\n","🎯 Validation du gel du backbone:\n"," • Statut global: ✅ OK\n"," • Backbone correctement gelé (27.7% ≥ 20%)\n"," • Plage attendue pour mobilenetv2: 20-35%\n","\n","✅ Analyse terminée avec succès\n","\n","======================================================================\n","🏗️ ANALYSE COMPLÈTE: U-Net_mobilenetv2_frozen_augFalse\n","======================================================================\n","📊 Configuration:\n"," • Architecture: U-Net\n"," • Backbone: mobilenetv2\n"," • Encoder gelé: ✅ Oui\n"," • Type: Frozen Pretrained\n"," • Augmentation: ❌ Non\n"," • Learning rate: 0.0001\n"," • Époques prévues: 20\n"," • Taille images: (224, 224)\n","\n","🔧 Création du modèle...\n","\n","🔢 Paramètres:\n"," • Total: 8,050,051\n"," • Entraînables: 5,824,195 (72.3%)\n"," • Gelés: 2,225,856 (27.7%)\n","\n","🔍 TOUTES LES COUCHES (🟢 = Entraînable, 🔴 = Gelé):\n","----------------------------------------------------------------------\n"," 1. 🔴 input_layer 0 params - GELÉ\n"," 2. 🔴 Conv1_pad 0 params - GELÉ\n"," 3. 🔴 Conv1 864 params - GELÉ\n"," 4. 🟢 bn_Conv1 128 params - ENTRAÎNABLE\n"," 5. 🔴 Conv1_relu 0 params - GELÉ\n"," 6. 🔴 expanded_conv_depthwise 288 params - GELÉ\n"," 7. 🟢 expanded_conv_depthwise_BN 128 params - ENTRAÎNABLE\n"," 8. 🔴 expanded_conv_depthwise_relu 0 params - GELÉ\n"," 9. 🔴 expanded_conv_project 512 params - GELÉ\n","10. 🟢 expanded_conv_project_BN 64 params - ENTRAÎNABLE\n","11. 🔴 block_1_expand 1,536 params - GELÉ\n","12. 🟢 block_1_expand_BN 384 params - ENTRAÎNABLE\n","13. 🔴 block_1_expand_relu 0 params - GELÉ\n","14. 🔴 block_1_pad 0 params - GELÉ\n","15. 🔴 block_1_depthwise 864 params - GELÉ\n","16. 🟢 block_1_depthwise_BN 384 params - ENTRAÎNABLE\n","17. 🔴 block_1_depthwise_relu 0 params - GELÉ\n","18. 🔴 block_1_project 2,304 params - GELÉ\n","19. 🟢 block_1_project_BN 96 params - ENTRAÎNABLE\n","20. 🔴 block_2_expand 3,456 params - GELÉ\n","21. 🟢 block_2_expand_BN 576 params - ENTRAÎNABLE\n","22. 🔴 block_2_expand_relu 0 params - GELÉ\n","23. 🔴 block_2_depthwise 1,296 params - GELÉ\n","24. 🟢 block_2_depthwise_BN 576 params - ENTRAÎNABLE\n","25. 🔴 block_2_depthwise_relu 0 params - GELÉ\n","26. 🔴 block_2_project 3,456 params - GELÉ\n","27. 🟢 block_2_project_BN 96 params - ENTRAÎNABLE\n","28. 🔴 block_2_add 0 params - GELÉ\n","29. 🔴 block_3_expand 3,456 params - GELÉ\n","30. 🟢 block_3_expand_BN 576 params - ENTRAÎNABLE\n","31. 🔴 block_3_expand_relu 0 params - GELÉ\n","32. 🔴 block_3_pad 0 params - GELÉ\n","33. 🔴 block_3_depthwise 1,296 params - GELÉ\n","34. 🟢 block_3_depthwise_BN 576 params - ENTRAÎNABLE\n","35. 🔴 block_3_depthwise_relu 0 params - GELÉ\n","36. 🔴 block_3_project 4,608 params - GELÉ\n","37. 🟢 block_3_project_BN 128 params - ENTRAÎNABLE\n","38. 🔴 block_4_expand 6,144 params - GELÉ\n","39. 🟢 block_4_expand_BN 768 params - ENTRAÎNABLE\n","40. 🔴 block_4_expand_relu 0 params - GELÉ\n","41. 🔴 block_4_depthwise 1,728 params - GELÉ\n","42. 🟢 block_4_depthwise_BN 768 params - ENTRAÎNABLE\n","43. 🔴 block_4_depthwise_relu 0 params - GELÉ\n","44. 🔴 block_4_project 6,144 params - GELÉ\n","45. 🟢 block_4_project_BN 128 params - ENTRAÎNABLE\n","46. 🔴 block_4_add 0 params - GELÉ\n","47. 🔴 block_5_expand 6,144 params - GELÉ\n","48. 🟢 block_5_expand_BN 768 params - ENTRAÎNABLE\n","49. 🔴 block_5_expand_relu 0 params - GELÉ\n","50. 🔴 block_5_depthwise 1,728 params - GELÉ\n","51. 🟢 block_5_depthwise_BN 768 params - ENTRAÎNABLE\n","52. 🔴 block_5_depthwise_relu 0 params - GELÉ\n","53. 🔴 block_5_project 6,144 params - GELÉ\n","54. 🟢 block_5_project_BN 128 params - ENTRAÎNABLE\n","55. 🔴 block_5_add 0 params - GELÉ\n","56. 🔴 block_6_expand 6,144 params - GELÉ\n","57. 🟢 block_6_expand_BN 768 params - ENTRAÎNABLE\n","58. 🔴 block_6_expand_relu 0 params - GELÉ\n","59. 🔴 block_6_pad 0 params - GELÉ\n","60. 🔴 block_6_depthwise 1,728 params - GELÉ\n","61. 🟢 block_6_depthwise_BN 768 params - ENTRAÎNABLE\n","62. 🔴 block_6_depthwise_relu 0 params - GELÉ\n","63. 🔴 block_6_project 12,288 params - GELÉ\n","64. 🟢 block_6_project_BN 256 params - ENTRAÎNABLE\n","65. 🔴 block_7_expand 24,576 params - GELÉ\n","66. 🟢 block_7_expand_BN 1,536 params - ENTRAÎNABLE\n","67. 🔴 block_7_expand_relu 0 params - GELÉ\n","68. 🔴 block_7_depthwise 3,456 params - GELÉ\n","69. 🟢 block_7_depthwise_BN 1,536 params - ENTRAÎNABLE\n","70. 🔴 block_7_depthwise_relu 0 params - GELÉ\n","71. 🔴 block_7_project 24,576 params - GELÉ\n","72. 🟢 block_7_project_BN 256 params - ENTRAÎNABLE\n","73. 🔴 block_7_add 0 params - GELÉ\n","74. 🔴 block_8_expand 24,576 params - GELÉ\n","75. 🟢 block_8_expand_BN 1,536 params - ENTRAÎNABLE\n","76. 🔴 block_8_expand_relu 0 params - GELÉ\n","77. 🔴 block_8_depthwise 3,456 params - GELÉ\n","78. 🟢 block_8_depthwise_BN 1,536 params - ENTRAÎNABLE\n","79. 🔴 block_8_depthwise_relu 0 params - GELÉ\n","80. 🔴 block_8_project 24,576 params - GELÉ\n","81. 🟢 block_8_project_BN 256 params - ENTRAÎNABLE\n","82. 🔴 block_8_add 0 params - GELÉ\n","83. 🔴 block_9_expand 24,576 params - GELÉ\n","84. 🟢 block_9_expand_BN 1,536 params - ENTRAÎNABLE\n","85. 🔴 block_9_expand_relu 0 params - GELÉ\n","86. 🔴 block_9_depthwise 3,456 params - GELÉ\n","87. 🟢 block_9_depthwise_BN 1,536 params - ENTRAÎNABLE\n","88. 🔴 block_9_depthwise_relu 0 params - GELÉ\n","89. 🔴 block_9_project 24,576 params - GELÉ\n","90. 🟢 block_9_project_BN 256 params - ENTRAÎNABLE\n","91. 🔴 block_9_add 0 params - GELÉ\n","92. 🔴 block_10_expand 24,576 params - GELÉ\n","93. 🟢 block_10_expand_BN 1,536 params - ENTRAÎNABLE\n","94. 🔴 block_10_expand_relu 0 params - GELÉ\n","95. 🔴 block_10_depthwise 3,456 params - GELÉ\n","96. 🟢 block_10_depthwise_BN 1,536 params - ENTRAÎNABLE\n","97. 🔴 block_10_depthwise_relu 0 params - GELÉ\n","98. 🔴 block_10_project 36,864 params - GELÉ\n","99. 🟢 block_10_project_BN 384 params - ENTRAÎNABLE\n","100. 🔴 block_11_expand 55,296 params - GELÉ\n","101. 🟢 block_11_expand_BN 2,304 params - ENTRAÎNABLE\n","102. 🔴 block_11_expand_relu 0 params - GELÉ\n","103. 🔴 block_11_depthwise 5,184 params - GELÉ\n","104. 🟢 block_11_depthwise_BN 2,304 params - ENTRAÎNABLE\n","105. 🔴 block_11_depthwise_relu 0 params - GELÉ\n","106. 🔴 block_11_project 55,296 params - GELÉ\n","107. 🟢 block_11_project_BN 384 params - ENTRAÎNABLE\n","108. 🔴 block_11_add 0 params - GELÉ\n","109. 🔴 block_12_expand 55,296 params - GELÉ\n","110. 🟢 block_12_expand_BN 2,304 params - ENTRAÎNABLE\n","111. 🔴 block_12_expand_relu 0 params - GELÉ\n","112. 🔴 block_12_depthwise 5,184 params - GELÉ\n","113. 🟢 block_12_depthwise_BN 2,304 params - ENTRAÎNABLE\n","114. 🔴 block_12_depthwise_relu 0 params - GELÉ\n","115. 🔴 block_12_project 55,296 params - GELÉ\n","116. 🟢 block_12_project_BN 384 params - ENTRAÎNABLE\n","117. 🔴 block_12_add 0 params - GELÉ\n","118. 🔴 block_13_expand 55,296 params - GELÉ\n","119. 🟢 block_13_expand_BN 2,304 params - ENTRAÎNABLE\n","120. 🔴 block_13_expand_relu 0 params - GELÉ\n","121. 🔴 block_13_pad 0 params - GELÉ\n","122. 🔴 block_13_depthwise 5,184 params - GELÉ\n","123. 🟢 block_13_depthwise_BN 2,304 params - ENTRAÎNABLE\n","124. 🔴 block_13_depthwise_relu 0 params - GELÉ\n","125. 🔴 block_13_project 92,160 params - GELÉ\n","126. 🟢 block_13_project_BN 640 params - ENTRAÎNABLE\n","127. 🔴 block_14_expand 153,600 params - GELÉ\n","128. 🟢 block_14_expand_BN 3,840 params - ENTRAÎNABLE\n","129. 🔴 block_14_expand_relu 0 params - GELÉ\n","130. 🔴 block_14_depthwise 8,640 params - GELÉ\n","131. 🟢 block_14_depthwise_BN 3,840 params - ENTRAÎNABLE\n","132. 🔴 block_14_depthwise_relu 0 params - GELÉ\n","133. 🔴 block_14_project 153,600 params - GELÉ\n","134. 🟢 block_14_project_BN 640 params - ENTRAÎNABLE\n","135. 🔴 block_14_add 0 params - GELÉ\n","136. 🔴 block_15_expand 153,600 params - GELÉ\n","137. 🟢 block_15_expand_BN 3,840 params - ENTRAÎNABLE\n","138. 🔴 block_15_expand_relu 0 params - GELÉ\n","139. 🔴 block_15_depthwise 8,640 params - GELÉ\n","140. 🟢 block_15_depthwise_BN 3,840 params - ENTRAÎNABLE\n","141. 🔴 block_15_depthwise_relu 0 params - GELÉ\n","142. 🔴 block_15_project 153,600 params - GELÉ\n","143. 🟢 block_15_project_BN 640 params - ENTRAÎNABLE\n","144. 🔴 block_15_add 0 params - GELÉ\n","145. 🔴 block_16_expand 153,600 params - GELÉ\n","146. 🟢 block_16_expand_BN 3,840 params - ENTRAÎNABLE\n","147. 🔴 block_16_expand_relu 0 params - GELÉ\n","148. 🔴 block_16_depthwise 8,640 params - GELÉ\n","149. 🟢 block_16_depthwise_BN 3,840 params - ENTRAÎNABLE\n","150. 🔴 block_16_depthwise_relu 0 params - GELÉ\n","151. 🔴 block_16_project 307,200 params - GELÉ\n","152. 🟢 block_16_project_BN 1,280 params - ENTRAÎNABLE\n","153. 🔴 Conv_1 409,600 params - GELÉ\n","154. 🟢 Conv_1_bn 5,120 params - ENTRAÎNABLE\n","155. 🔴 out_relu 0 params - GELÉ\n","156. 🟢 decoder_stage0_upsampling 0 params - ENTRAÎNABLE (DECODER)\n","157. 🟢 decoder_stage0_concat 0 params - ENTRAÎNABLE (DECODER)\n","158. 🟢 decoder_stage0a_conv 4,276,224 params - ENTRAÎNABLE (DECODER)\n","159. 🟢 decoder_stage0a_bn 1,024 params - ENTRAÎNABLE (DECODER)\n","160. 🟢 decoder_stage0a_relu 0 params - ENTRAÎNABLE (DECODER)\n","161. 🟢 decoder_stage0b_conv 589,824 params - ENTRAÎNABLE (DECODER)\n","162. 🟢 decoder_stage0b_bn 1,024 params - ENTRAÎNABLE (DECODER)\n","163. 🟢 decoder_stage0b_relu 0 params - ENTRAÎNABLE (DECODER)\n","164. 🟢 decoder_stage1_upsampling 0 params - ENTRAÎNABLE (DECODER)\n","165. 🟢 decoder_stage1_concat 0 params - ENTRAÎNABLE (DECODER)\n","166. 🟢 decoder_stage1a_conv 516,096 params - ENTRAÎNABLE (DECODER)\n","167. 🟢 decoder_stage1a_bn 512 params - ENTRAÎNABLE (DECODER)\n","168. 🟢 decoder_stage1a_relu 0 params - ENTRAÎNABLE (DECODER)\n","169. 🟢 decoder_stage1b_conv 147,456 params - ENTRAÎNABLE (DECODER)\n","170. 🟢 decoder_stage1b_bn 512 params - ENTRAÎNABLE (DECODER)\n","171. 🟢 decoder_stage1b_relu 0 params - ENTRAÎNABLE (DECODER)\n","172. 🟢 decoder_stage2_upsampling 0 params - ENTRAÎNABLE (DECODER)\n","173. 🟢 decoder_stage2_concat 0 params - ENTRAÎNABLE (DECODER)\n","174. 🟢 decoder_stage2a_conv 156,672 params - ENTRAÎNABLE (DECODER)\n","175. 🟢 decoder_stage2a_bn 256 params - ENTRAÎNABLE (DECODER)\n","176. 🟢 decoder_stage2a_relu 0 params - ENTRAÎNABLE (DECODER)\n","177. 🟢 decoder_stage2b_conv 36,864 params - ENTRAÎNABLE (DECODER)\n","178. 🟢 decoder_stage2b_bn 256 params - ENTRAÎNABLE (DECODER)\n","179. 🟢 decoder_stage2b_relu 0 params - ENTRAÎNABLE (DECODER)\n","180. 🟢 decoder_stage3_upsampling 0 params - ENTRAÎNABLE (DECODER)\n","181. 🟢 decoder_stage3_concat 0 params - ENTRAÎNABLE (DECODER)\n","182. 🟢 decoder_stage3a_conv 46,080 params - ENTRAÎNABLE (DECODER)\n","183. 🟢 decoder_stage3a_bn 128 params - ENTRAÎNABLE (DECODER)\n","184. 🟢 decoder_stage3a_relu 0 params - ENTRAÎNABLE (DECODER)\n","185. 🟢 decoder_stage3b_conv 9,216 params - ENTRAÎNABLE (DECODER)\n","186. 🟢 decoder_stage3b_bn 128 params - ENTRAÎNABLE (DECODER)\n","187. 🟢 decoder_stage3b_relu 0 params - ENTRAÎNABLE (DECODER)\n","188. 🟢 decoder_stage4_upsampling 0 params - ENTRAÎNABLE (DECODER)\n","189. 🟢 decoder_stage4a_conv 4,608 params - ENTRAÎNABLE (DECODER)\n","190. 🟢 decoder_stage4a_bn 64 params - ENTRAÎNABLE (DECODER)\n","191. 🟢 decoder_stage4a_relu 0 params - ENTRAÎNABLE (DECODER)\n","192. 🟢 decoder_stage4b_conv 2,304 params - ENTRAÎNABLE (DECODER)\n","193. 🟢 decoder_stage4b_bn 64 params - ENTRAÎNABLE (DECODER)\n","194. 🟢 decoder_stage4b_relu 0 params - ENTRAÎNABLE (DECODER)\n","195. 🟢 final_conv 2,755 params - ENTRAÎNABLE (CLASSIFICATION)\n","196. 🟢 softmax 0 params - ENTRAÎNABLE\n","\n","🎯 Validation du gel du backbone:\n"," • Statut global: ✅ OK\n"," • Backbone correctement gelé (27.7% ≥ 20%)\n"," • Plage attendue pour mobilenetv2: 20-35%\n","\n","✅ Analyse terminée avec succès\n","\n","🔍 COUCHES - U-Net_mobilenetv2_frozen_augFalse\n","======================================================================\n","🟢 ENTRAÎNABLE | 🔴 GELÉ\n","----------------------------------------------------------------------\n"," 1. 🔴 input_layer 0 AUTRE\n"," 2. 🔴 Conv1_pad 0 AUTRE\n"," 3. 🔴 Conv1 864 AUTRE\n"," 4. 🟢 bn_Conv1 128 AUTRE\n"," 5. 🔴 Conv1_relu 0 AUTRE\n"," 6. 🔴 expanded_conv_depthwise 288 AUTRE\n"," 7. 🟢 expanded_conv_depthwise_BN 128 AUTRE\n"," 8. 🔴 expanded_conv_depthwise_relu 0 AUTRE\n"," 9. 🔴 expanded_conv_project 512 AUTRE\n","10. 🟢 expanded_conv_project_BN 64 AUTRE\n","11. 🔴 block_1_expand 1,536 AUTRE\n","12. 🟢 block_1_expand_BN 384 AUTRE\n","13. 🔴 block_1_expand_relu 0 AUTRE\n","14. 🔴 block_1_pad 0 AUTRE\n","15. 🔴 block_1_depthwise 864 AUTRE\n","16. 🟢 block_1_depthwise_BN 384 AUTRE\n","17. 🔴 block_1_depthwise_relu 0 AUTRE\n","18. 🔴 block_1_project 2,304 AUTRE\n","19. 🟢 block_1_project_BN 96 AUTRE\n","20. 🔴 block_2_expand 3,456 AUTRE\n","21. 🟢 block_2_expand_BN 576 AUTRE\n","22. 🔴 block_2_expand_relu 0 AUTRE\n","23. 🔴 block_2_depthwise 1,296 AUTRE\n","24. 🟢 block_2_depthwise_BN 576 AUTRE\n","25. 🔴 block_2_depthwise_relu 0 AUTRE\n","26. 🔴 block_2_project 3,456 AUTRE\n","27. 🟢 block_2_project_BN 96 AUTRE\n","28. 🔴 block_2_add 0 AUTRE\n","29. 🔴 block_3_expand 3,456 AUTRE\n","30. 🟢 block_3_expand_BN 576 AUTRE\n","31. 🔴 block_3_expand_relu 0 AUTRE\n","32. 🔴 block_3_pad 0 AUTRE\n","33. 🔴 block_3_depthwise 1,296 AUTRE\n","34. 🟢 block_3_depthwise_BN 576 AUTRE\n","35. 🔴 block_3_depthwise_relu 0 AUTRE\n","36. 🔴 block_3_project 4,608 AUTRE\n","37. 🟢 block_3_project_BN 128 AUTRE\n","38. 🔴 block_4_expand 6,144 AUTRE\n","39. 🟢 block_4_expand_BN 768 AUTRE\n","40. 🔴 block_4_expand_relu 0 AUTRE\n","41. 🔴 block_4_depthwise 1,728 AUTRE\n","42. 🟢 block_4_depthwise_BN 768 AUTRE\n","43. 🔴 block_4_depthwise_relu 0 AUTRE\n","44. 🔴 block_4_project 6,144 AUTRE\n","45. 🟢 block_4_project_BN 128 AUTRE\n","46. 🔴 block_4_add 0 AUTRE\n","47. 🔴 block_5_expand 6,144 AUTRE\n","48. 🟢 block_5_expand_BN 768 AUTRE\n","49. 🔴 block_5_expand_relu 0 AUTRE\n","50. 🔴 block_5_depthwise 1,728 AUTRE\n","51. 🟢 block_5_depthwise_BN 768 AUTRE\n","52. 🔴 block_5_depthwise_relu 0 AUTRE\n","53. 🔴 block_5_project 6,144 AUTRE\n","54. 🟢 block_5_project_BN 128 AUTRE\n","55. 🔴 block_5_add 0 AUTRE\n","56. 🔴 block_6_expand 6,144 AUTRE\n","57. 🟢 block_6_expand_BN 768 AUTRE\n","58. 🔴 block_6_expand_relu 0 AUTRE\n","59. 🔴 block_6_pad 0 AUTRE\n","60. 🔴 block_6_depthwise 1,728 AUTRE\n","61. 🟢 block_6_depthwise_BN 768 AUTRE\n","62. 🔴 block_6_depthwise_relu 0 AUTRE\n","63. 🔴 block_6_project 12,288 AUTRE\n","64. 🟢 block_6_project_BN 256 AUTRE\n","65. 🔴 block_7_expand 24,576 AUTRE\n","66. 🟢 block_7_expand_BN 1,536 AUTRE\n","67. 🔴 block_7_expand_relu 0 AUTRE\n","68. 🔴 block_7_depthwise 3,456 AUTRE\n","69. 🟢 block_7_depthwise_BN 1,536 AUTRE\n","70. 🔴 block_7_depthwise_relu 0 AUTRE\n","71. 🔴 block_7_project 24,576 AUTRE\n","72. 🟢 block_7_project_BN 256 AUTRE\n","73. 🔴 block_7_add 0 AUTRE\n","74. 🔴 block_8_expand 24,576 AUTRE\n","75. 🟢 block_8_expand_BN 1,536 AUTRE\n","76. 🔴 block_8_expand_relu 0 AUTRE\n","77. 🔴 block_8_depthwise 3,456 AUTRE\n","78. 🟢 block_8_depthwise_BN 1,536 AUTRE\n","79. 🔴 block_8_depthwise_relu 0 AUTRE\n","80. 🔴 block_8_project 24,576 AUTRE\n","81. 🟢 block_8_project_BN 256 AUTRE\n","82. 🔴 block_8_add 0 AUTRE\n","83. 🔴 block_9_expand 24,576 AUTRE\n","84. 🟢 block_9_expand_BN 1,536 AUTRE\n","85. 🔴 block_9_expand_relu 0 AUTRE\n","86. 🔴 block_9_depthwise 3,456 AUTRE\n","87. 🟢 block_9_depthwise_BN 1,536 AUTRE\n","88. 🔴 block_9_depthwise_relu 0 AUTRE\n","89. 🔴 block_9_project 24,576 AUTRE\n","90. 🟢 block_9_project_BN 256 AUTRE\n","91. 🔴 block_9_add 0 AUTRE\n","92. 🔴 block_10_expand 24,576 AUTRE\n","93. 🟢 block_10_expand_BN 1,536 AUTRE\n","94. 🔴 block_10_expand_relu 0 AUTRE\n","95. 🔴 block_10_depthwise 3,456 AUTRE\n","96. 🟢 block_10_depthwise_BN 1,536 AUTRE\n","97. 🔴 block_10_depthwise_relu 0 AUTRE\n","98. 🔴 block_10_project 36,864 AUTRE\n","99. 🟢 block_10_project_BN 384 AUTRE\n","100. 🔴 block_11_expand 55,296 AUTRE\n","101. 🟢 block_11_expand_BN 2,304 AUTRE\n","102. 🔴 block_11_expand_relu 0 AUTRE\n","103. 🔴 block_11_depthwise 5,184 AUTRE\n","104. 🟢 block_11_depthwise_BN 2,304 AUTRE\n","105. 🔴 block_11_depthwise_relu 0 AUTRE\n","106. 🔴 block_11_project 55,296 AUTRE\n","107. 🟢 block_11_project_BN 384 AUTRE\n","108. 🔴 block_11_add 0 AUTRE\n","109. 🔴 block_12_expand 55,296 AUTRE\n","110. 🟢 block_12_expand_BN 2,304 AUTRE\n","111. 🔴 block_12_expand_relu 0 AUTRE\n","112. 🔴 block_12_depthwise 5,184 AUTRE\n","113. 🟢 block_12_depthwise_BN 2,304 AUTRE\n","114. 🔴 block_12_depthwise_relu 0 AUTRE\n","115. 🔴 block_12_project 55,296 AUTRE\n","116. 🟢 block_12_project_BN 384 AUTRE\n","117. 🔴 block_12_add 0 AUTRE\n","118. 🔴 block_13_expand 55,296 AUTRE\n","119. 🟢 block_13_expand_BN 2,304 AUTRE\n","120. 🔴 block_13_expand_relu 0 AUTRE\n","121. 🔴 block_13_pad 0 AUTRE\n","122. 🔴 block_13_depthwise 5,184 AUTRE\n","123. 🟢 block_13_depthwise_BN 2,304 AUTRE\n","124. 🔴 block_13_depthwise_relu 0 AUTRE\n","125. 🔴 block_13_project 92,160 AUTRE\n","126. 🟢 block_13_project_BN 640 AUTRE\n","127. 🔴 block_14_expand 153,600 AUTRE\n","128. 🟢 block_14_expand_BN 3,840 AUTRE\n","129. 🔴 block_14_expand_relu 0 AUTRE\n","130. 🔴 block_14_depthwise 8,640 AUTRE\n","131. 🟢 block_14_depthwise_BN 3,840 AUTRE\n","132. 🔴 block_14_depthwise_relu 0 AUTRE\n","133. 🔴 block_14_project 153,600 AUTRE\n","134. 🟢 block_14_project_BN 640 AUTRE\n","135. 🔴 block_14_add 0 AUTRE\n","136. 🔴 block_15_expand 153,600 AUTRE\n","137. 🟢 block_15_expand_BN 3,840 AUTRE\n","138. 🔴 block_15_expand_relu 0 AUTRE\n","139. 🔴 block_15_depthwise 8,640 AUTRE\n","140. 🟢 block_15_depthwise_BN 3,840 AUTRE\n","141. 🔴 block_15_depthwise_relu 0 AUTRE\n","142. 🔴 block_15_project 153,600 AUTRE\n","143. 🟢 block_15_project_BN 640 AUTRE\n","144. 🔴 block_15_add 0 AUTRE\n","145. 🔴 block_16_expand 153,600 AUTRE\n","146. 🟢 block_16_expand_BN 3,840 AUTRE\n","147. 🔴 block_16_expand_relu 0 AUTRE\n","148. 🔴 block_16_depthwise 8,640 AUTRE\n","149. 🟢 block_16_depthwise_BN 3,840 AUTRE\n","150. 🔴 block_16_depthwise_relu 0 AUTRE\n","151. 🔴 block_16_project 307,200 AUTRE\n","152. 🟢 block_16_project_BN 1,280 AUTRE\n","153. 🔴 Conv_1 409,600 AUTRE\n","154. 🟢 Conv_1_bn 5,120 AUTRE\n","155. 🔴 out_relu 0 AUTRE\n","156. 🟢 decoder_stage0_upsampling 0 DECODER\n","157. 🟢 decoder_stage0_concat 0 DECODER\n","158. 🟢 decoder_stage0a_conv 4,276,224 DECODER\n","159. 🟢 decoder_stage0a_bn 1,024 DECODER\n","160. 🟢 decoder_stage0a_relu 0 DECODER\n","161. 🟢 decoder_stage0b_conv 589,824 DECODER\n","162. 🟢 decoder_stage0b_bn 1,024 DECODER\n","163. 🟢 decoder_stage0b_relu 0 DECODER\n","164. 🟢 decoder_stage1_upsampling 0 DECODER\n","165. 🟢 decoder_stage1_concat 0 DECODER\n","166. 🟢 decoder_stage1a_conv 516,096 DECODER\n","167. 🟢 decoder_stage1a_bn 512 DECODER\n","168. 🟢 decoder_stage1a_relu 0 DECODER\n","169. 🟢 decoder_stage1b_conv 147,456 DECODER\n","170. 🟢 decoder_stage1b_bn 512 DECODER\n","171. 🟢 decoder_stage1b_relu 0 DECODER\n","172. 🟢 decoder_stage2_upsampling 0 DECODER\n","173. 🟢 decoder_stage2_concat 0 DECODER\n","174. 🟢 decoder_stage2a_conv 156,672 DECODER\n","175. 🟢 decoder_stage2a_bn 256 DECODER\n","176. 🟢 decoder_stage2a_relu 0 DECODER\n","177. 🟢 decoder_stage2b_conv 36,864 DECODER\n","178. 🟢 decoder_stage2b_bn 256 DECODER\n","179. 🟢 decoder_stage2b_relu 0 DECODER\n","180. 🟢 decoder_stage3_upsampling 0 DECODER\n","181. 🟢 decoder_stage3_concat 0 DECODER\n","182. 🟢 decoder_stage3a_conv 46,080 DECODER\n","183. 🟢 decoder_stage3a_bn 128 DECODER\n","184. 🟢 decoder_stage3a_relu 0 DECODER\n","185. 🟢 decoder_stage3b_conv 9,216 DECODER\n","186. 🟢 decoder_stage3b_bn 128 DECODER\n","187. 🟢 decoder_stage3b_relu 0 DECODER\n","188. 🟢 decoder_stage4_upsampling 0 DECODER\n","189. 🟢 decoder_stage4a_conv 4,608 DECODER\n","190. 🟢 decoder_stage4a_bn 64 DECODER\n","191. 🟢 decoder_stage4a_relu 0 DECODER\n","192. 🟢 decoder_stage4b_conv 2,304 DECODER\n","193. 🟢 decoder_stage4b_bn 64 DECODER\n","194. 🟢 decoder_stage4b_relu 0 DECODER\n","195. 🟢 final_conv 2,755 AUTRE\n","196. 🟢 softmax 0 AUTRE\n","----------------------------------------------------------------------\n","📊 RÉSUMÉ: 93 entraînables, 103 gelés\n"]}],"source":["# Analyse standard (backbone + résumé)\n","analyze_model_comprehensive(experimental_configurations[4], num_classes=19)\n","\n","# Analyse complète (toutes les couches)\n","analyze_model_comprehensive(experimental_configurations[4], num_classes=19, detailed_layers=True)\n","\n","# Vue compacte des couches uniquement\n","show_all_layers_status(experimental_configurations[4], num_classes=19)"]},{"cell_type":"markdown","metadata":{"id":"yGNRUQdxT4xL"},"source":["## Préparation des données"]},{"cell_type":"markdown","metadata":{"id":"e5mbV7CtEfoS"},"source":["Ici, les données vont être prétraitées et des datasets d'entrainement, de validation et de test vont être constitués.\n","Il est important de noter la correspondance terminologique entre le jeu de données officiel et les jeux de données que nous constituerons dans le cadre de cet entraînement :\\\n","• Split 'train' Cityscapes → Train (80%) + Validation (20%)\\\n","• Split 'val' Cityscapes → Test final\\\n","• Split 'test' Cityscapes → non utilisé\n","\n","En effet, le dossier \"test\" de Cityscapes ne contient pas de masque de vérité et sert à effectuer un benchmark de performances dans le cadre d'un concours Kaggel (résultats cachés, donc)."]},{"cell_type":"code","execution_count":23,"metadata":{"id":"wMEw3js7CNZE","executionInfo":{"status":"ok","timestamp":1755000408727,"user_tz":-120,"elapsed":21,"user":{"displayName":"Grégoire Mureau","userId":"04855910922109670788"}}},"outputs":[],"source":["def prepare_cityscapes_dataset(\n"," cityscapes_dir,\n"," split='train',\n"," validation_split=0.2,\n"," batch_size=8,\n"," img_size=(224, 224),\n"," encoder=None,\n"," id_to_group_mapping=None\n","):\n"," \"\"\"\n"," Prépare les datasets Cityscapes avec prétraitement adapté au backbone\n","\n"," Args:\n"," cityscapes_dir (str): Chemin vers le répertoire Cityscapes\n"," split (str): Split Cityscapes à utiliser ('train', 'val', 'test')\n"," validation_split (float): Si split='train', proportion à garder pour validation\n"," batch_size (int): Taille des batches\n"," img_size (tuple): Taille des images (H, W)\n"," encoder (str): Nom du backbone pour preprocessing ou None\n"," id_to_group_mapping (list ou np.ndarray): Mapping des IDs de classes ou None\n","\n"," Returns:\n"," Dataset(s): Si split='train' et validation_split > 0 → (train_dataset, val_dataset)\n"," Sinon → dataset unique\n"," \"\"\"\n"," if not os.path.exists(cityscapes_dir):\n"," raise ValueError(f\"Le répertoire Cityscapes n'existe pas: {cityscapes_dir}\")\n","\n"," images_dir = os.path.join(cityscapes_dir, 'leftImg8bit', split)\n"," masks_dir = os.path.join(cityscapes_dir, 'gtFine', split)\n","\n"," if not os.path.exists(images_dir) or not os.path.exists(masks_dir):\n"," raise ValueError(f\"Split Cityscapes '{split}' non trouvé dans {cityscapes_dir}\")\n","\n"," image_paths = []\n"," mask_paths = []\n","\n"," print(f\"🔍 Lecture du split Cityscapes '{split}'...\")\n","\n"," for city in os.listdir(images_dir):\n"," city_img_dir = os.path.join(images_dir, city)\n"," city_mask_dir = os.path.join(masks_dir, city)\n","\n"," if os.path.isdir(city_img_dir) and os.path.isdir(city_mask_dir):\n"," for file_name in os.listdir(city_img_dir):\n"," if file_name.endswith('_leftImg8bit.png'):\n"," img_path = os.path.join(city_img_dir, file_name)\n"," mask_name = file_name.replace('_leftImg8bit.png', '_gtFine_labelIds.png')\n"," mask_path = os.path.join(city_mask_dir, mask_name)\n"," if os.path.exists(img_path) and os.path.exists(mask_path):\n"," image_paths.append(img_path)\n"," mask_paths.append(mask_path)\n"," else:\n"," print(f\"⚠️ Fichier manquant pour {file_name}\")\n","\n"," print(f\"📊 Split Cityscapes '{split}': {len(image_paths)} images trouvées\")\n","\n"," if len(image_paths) == 0:\n"," raise ValueError(f\"Aucun fichier valide trouvé pour le split '{split}'!\")\n","\n"," if split == 'train' and validation_split > 0:\n"," print(f\"✂️ Découpage en Train/Validation ({int((1-validation_split)*100)}%/{int(validation_split*100)}%)\")\n","\n"," combined = list(zip(image_paths, mask_paths))\n"," random.seed(42)\n"," random.shuffle(combined)\n"," image_paths, mask_paths = zip(*combined)\n","\n"," total_samples = len(image_paths)\n"," val_samples = int(total_samples * validation_split)\n"," train_samples = total_samples - val_samples\n","\n"," train_img_paths = list(image_paths[:train_samples])\n"," train_mask_paths = list(mask_paths[:train_samples])\n"," val_img_paths = list(image_paths[train_samples:])\n"," val_mask_paths = list(mask_paths[train_samples:])\n","\n"," print(f\" └─ 📚 Train final: {len(train_img_paths)} images\")\n"," print(f\" └─ 📊 Validation final: {len(val_img_paths)} images\")\n"," else:\n"," usage = \"Test\" if split == 'val' else split.title()\n"," print(f\"📋 Utilisation complète comme dataset de {usage}\")\n","\n"," train_img_paths, train_mask_paths = image_paths, mask_paths\n"," val_img_paths, val_mask_paths = [], []\n","\n"," preprocessing_fn = get_preprocessing_fn(encoder)\n","\n"," def load_and_preprocess(img_path, mask_path):\n"," img = tf.io.read_file(img_path)\n"," img = tf.image.decode_png(img, channels=3)\n"," img = tf.image.resize(img, img_size, method='bilinear')\n"," img = tf.cast(img, tf.float32)\n"," img = preprocessing_fn(img)\n","\n"," mask = tf.io.read_file(mask_path)\n"," mask = tf.image.decode_png(mask, channels=1)\n"," mask = tf.image.resize(mask, img_size, method='nearest')\n"," mask = tf.cast(mask, tf.int32)\n","\n"," if id_to_group_mapping is not None:\n"," mask = map_mask_ids(mask, id_to_group_mapping)\n","\n"," mask = tf.squeeze(mask, axis=-1)\n"," return img, mask\n","\n"," def augment_data(image, mask):\n"," mask = tf.expand_dims(mask, axis=-1)\n","\n"," if tf.random.uniform(()) > 0.5:\n"," image = tf.image.flip_left_right(image)\n"," mask = tf.image.flip_left_right(mask)\n","\n"," image = tf.image.random_brightness(image, 0.1)\n"," image = tf.clip_by_value(image, 0, 1)\n","\n"," mask = tf.squeeze(mask, axis=-1)\n"," return image, mask\n","\n"," def create_dataset(image_paths, mask_paths, augment=False, shuffle_data=True):\n"," dataset = tf.data.Dataset.from_tensor_slices((image_paths, mask_paths))\n"," dataset = dataset.map(load_and_preprocess, num_parallel_calls=tf.data.AUTOTUNE)\n","\n"," if augment:\n"," dataset = dataset.map(augment_data, num_parallel_calls=tf.data.AUTOTUNE)\n","\n"," if shuffle_data:\n"," dataset = dataset.shuffle(buffer_size=1000)\n","\n"," dataset = dataset.batch(batch_size).prefetch(tf.data.AUTOTUNE)\n"," return dataset\n","\n"," if split == 'train' and validation_split > 0:\n"," train_dataset = create_dataset(train_img_paths, train_mask_paths, augment=True, shuffle_data=True)\n"," val_dataset = create_dataset(val_img_paths, val_mask_paths, augment=False, shuffle_data=False)\n"," return train_dataset, val_dataset\n"," else:\n"," dataset = create_dataset(train_img_paths, train_mask_paths, augment=False, shuffle_data=(split=='train'))\n"," return dataset\n"]},{"cell_type":"code","execution_count":24,"metadata":{"colab":{"base_uri":"https://localhost:8080/"},"executionInfo":{"elapsed":60423,"status":"ok","timestamp":1755000469150,"user":{"displayName":"Grégoire Mureau","userId":"04855910922109670788"},"user_tz":-120},"id":"VimLczDnCt7l","outputId":"e13908e6-ba98-4247-9884-6243aec44d75"},"outputs":[{"output_type":"stream","name":"stdout","text":["=== PRÉPARATION DES DATASETS CITYSCAPES ===\n","📊 Paramètres:\n"," • Taille image: (224, 224)\n"," • Taille batch: 8\n"," • Répertoire: /content/data/cityscapes_unzipped\n","\n","🗂️ Stratégie de découpage:\n"," • Split 'train' Cityscapes → Train (80%) + Validation (20%)\n"," • Split 'val' Cityscapes → Test final\n","\n","--- Découpage du split 'train' Cityscapes en Train/Validation ---\n","🔍 Lecture du split Cityscapes 'train'...\n","📊 Split Cityscapes 'train': 2975 images trouvées\n","✂️ Découpage en Train/Validation (80%/20%)\n"," └─ 📚 Train final: 2380 images\n"," └─ 📊 Validation final: 595 images\n","ℹ️ Modèle vanilla - pas de preprocessing\n","\n","--- Création du dataset de TEST à partir du split 'val' Cityscapes ---\n","🔍 Lecture du split Cityscapes 'val'...\n","📊 Split Cityscapes 'val': 500 images trouvées\n","📋 Utilisation complète comme dataset de Test\n","ℹ️ Modèle vanilla - pas de preprocessing\n","\n","✅ Datasets créés avec succès:\n"," 📚 Train: 2380 images (avec augmentation)\n"," 📊 Validation: 595 images (sans augmentation)\n"," 🧪 Test: 500 images (sans augmentation)\n","\n","📋 Récapitulatif:\n"," • Train dataset: pour l'entraînement des modèles\n"," • Validation dataset: pour la sélection des hyperparamètres\n"," • Test dataset: pour l'évaluation finale (données jamais vues)\n","\n","🔍 Vérification des formes:\n"," Train: images = (8, 224, 224, 3), masques = (8, 224, 224)\n"," Validation: images = (8, 224, 224, 3), masques = (8, 224, 224)\n"," Test: images = (8, 224, 224, 3), masques = (8, 224, 224)\n"]}],"source":["# === UTILISATION AVEC TERMINOLOGIE CLAIRE ===\n","# Définir les paramètres\n","BATCH_SIZE = 8\n","IMG_SIZE = (224, 224)\n","\n","# Vérifier si id_to_group existe, sinon définir None\n","try:\n"," id_mapping = id_to_group\n","except NameError:\n"," print(\"⚠️ Variable 'id_to_group' non définie, pas de mapping des IDs\")\n"," id_mapping = None\n","\n","print(\"=== PRÉPARATION DES DATASETS CITYSCAPES ===\")\n","print(f\"📊 Paramètres:\")\n","print(f\" • Taille image: {IMG_SIZE}\")\n","print(f\" • Taille batch: {BATCH_SIZE}\")\n","print(f\" • Répertoire: {cityscapes_dir}\")\n","\n","print(f\"\\n🗂️ Stratégie de découpage:\")\n","print(f\" • Split 'train' Cityscapes → Train (80%) + Validation (20%)\")\n","print(f\" • Split 'val' Cityscapes → Test final\")\n","\n","try:\n"," # 1. Créer TRAIN + VALIDATION à partir du split 'train' de Cityscapes\n"," print(f\"\\n--- Découpage du split 'train' Cityscapes en Train/Validation ---\")\n"," train_dataset, val_dataset = prepare_cityscapes_dataset(\n"," cityscapes_dir,\n"," split='train', # On utilise le split 'train' de Cityscapes\n"," batch_size=BATCH_SIZE,\n"," img_size=IMG_SIZE,\n"," validation_split=0.2, # 80% train, 20% validation\n"," id_to_group_mapping=id_mapping\n"," )\n","\n"," # 2. Créer TEST à partir du split 'val' de Cityscapes\n"," print(f\"\\n--- Création du dataset de TEST à partir du split 'val' Cityscapes ---\")\n"," test_dataset = prepare_cityscapes_dataset(\n"," cityscapes_dir,\n"," split='val', # On utilise le split 'val' de Cityscapes comme notre test\n"," batch_size=BATCH_SIZE,\n"," img_size=IMG_SIZE,\n"," id_to_group_mapping=id_mapping\n"," )\n","\n"," print(f\"\\n✅ Datasets créés avec succès:\")\n"," print(f\" 📚 Train: {2380} images (avec augmentation)\")\n"," print(f\" 📊 Validation: {595} images (sans augmentation)\")\n"," print(f\" 🧪 Test: {500} images (sans augmentation)\")\n","\n"," print(f\"\\n📋 Récapitulatif:\")\n"," print(f\" • Train dataset: pour l'entraînement des modèles\")\n"," print(f\" • Validation dataset: pour la sélection des hyperparamètres\")\n"," print(f\" • Test dataset: pour l'évaluation finale (données jamais vues)\")\n","\n"," # Vérification rapide\n"," print(f\"\\n🔍 Vérification des formes:\")\n"," for name, dataset in [(\"Train\", train_dataset), (\"Validation\", val_dataset), (\"Test\", test_dataset)]:\n"," sample = next(iter(dataset))\n"," print(f\" {name}: images = {sample[0].shape}, masques = {sample[1].shape}\")\n","\n","except Exception as e:\n"," print(f\"❌ Erreur: {e}\")\n"," print(\"Vérifications:\")\n"," print(\"1. Variable 'cityscapes_dir' définie ?\")\n"," print(\"2. Fichiers Cityscapes extraits ?\")\n"," print(\"3. Structure correcte (leftImg8bit/, gtFine/) ?\")\n"," raise"]},{"cell_type":"markdown","metadata":{"id":"q4W-fiAxSwWR"},"source":["## Visualisation des données"]},{"cell_type":"markdown","metadata":{"id":"aSz8Oa3sS3_n"},"source":["Une visualisation de quelques images (train, val, test) sera proposée avec les masques de vérité dont on propose une colorisation tab10 puis une colorisation plus spécifique"]},{"cell_type":"code","execution_count":25,"metadata":{"colab":{"base_uri":"https://localhost:8080/","height":1000,"output_embedded_package_id":"1oB3hgwKTFgjKC-6Yfdv8sOhQ7ktGVorA"},"executionInfo":{"elapsed":98553,"status":"ok","timestamp":1755000567721,"user":{"displayName":"Grégoire Mureau","userId":"04855910922109670788"},"user_tz":-120},"id":"mWO3yOl3Oq12","outputId":"4f1e59cc-22c3-4255-be28-7a43e3e8a22f"},"outputs":[{"output_type":"display_data","data":{"text/plain":"Output hidden; open in https://colab.research.google.com to view."},"metadata":{}}],"source":["def visualize_cityscapes_samples(\n"," datasets_dict=None,\n"," cityscapes_dir=None,\n"," num_samples=3,\n"," save_dir=None,\n"," show_stats=True,\n"," show_legend=True\n","):\n"," \"\"\"\n"," Visualise des échantillons Cityscapes avec options flexibles\n","\n"," Args:\n"," datasets_dict: {'train': train_dataset, 'val': val_dataset, 'test': test_dataset}\n"," cityscapes_dir: Chemin vers données brutes (si datasets_dict=None)\n"," num_samples: Nombre d'échantillons par dataset\n"," save_dir: Répertoire de sauvegarde (optionnel)\n"," show_stats: Afficher statistiques des classes\n"," show_legend: Créer légende des classes\n"," \"\"\"\n","\n"," # Mode 1: Depuis les datasets TensorFlow (préféré)\n"," if datasets_dict is not None:\n"," print(\"📊 Visualisation depuis les datasets TensorFlow...\")\n"," _visualize_from_datasets(datasets_dict, num_samples, save_dir, show_stats)\n","\n"," # Mode 2: Depuis les fichiers bruts (fallback)\n"," elif cityscapes_dir is not None:\n"," print(\"📁 Visualisation depuis les fichiers bruts...\")\n"," _visualize_from_files(cityscapes_dir, num_samples, save_dir, show_stats)\n","\n"," else:\n"," raise ValueError(\"Fournir soit datasets_dict soit cityscapes_dir\")\n","\n"," # Créer la légende si demandée\n"," if show_legend:\n"," _create_legend(save_dir)\n","\n","def _visualize_from_datasets(datasets_dict, num_samples, save_dir, show_stats):\n"," \"\"\"Visualise depuis les datasets TensorFlow (après preprocessing)\"\"\"\n","\n"," for split_name, dataset in datasets_dict.items():\n"," print(f\"\\n--- {split_name.title()} Dataset ---\")\n","\n"," sample_count = 0\n"," for batch in dataset.take(num_samples):\n"," images, masks = batch\n","\n"," # Prendre le premier échantillon du batch\n"," image = images[0]\n"," mask = masks[0]\n","\n"," # Créer la visualisation\n"," fig = _create_sample_plot(image.numpy(), mask.numpy(),\n"," f\"{split_name.title()} Sample {sample_count+1}\")\n","\n"," # Sauvegarder si demandé\n"," if save_dir:\n"," os.makedirs(save_dir, exist_ok=True)\n"," fig_path = os.path.join(save_dir, f'{split_name}_sample_{sample_count+1}.png')\n"," fig.savefig(fig_path, dpi=150, bbox_inches='tight')\n"," print(f\"💾 Sauvegardé: {fig_path}\")\n","\n"," plt.show()\n"," plt.close(fig)\n","\n"," # Afficher les stats si demandé\n"," if show_stats:\n"," _show_class_stats(mask.numpy())\n","\n"," sample_count += 1\n"," if sample_count >= num_samples:\n"," break\n","\n","def _visualize_from_files(cityscapes_dir, num_samples, save_dir, show_stats):\n"," \"\"\"Visualise depuis les fichiers bruts (fallback)\"\"\"\n","\n"," for split in ['train', 'val']:\n"," print(f\"\\n--- {split.title()} (Raw Files) ---\")\n","\n"," # Collecter les chemins (réutilise la logique de prepare_cityscapes_dataset)\n"," images_dir = os.path.join(cityscapes_dir, 'leftImg8bit', split)\n"," masks_dir = os.path.join(cityscapes_dir, 'gtFine', split)\n","\n"," if not os.path.exists(images_dir):\n"," print(f\"⚠️ Split '{split}' non trouvé, ignoré\")\n"," continue\n","\n"," image_paths, mask_paths = _collect_sample_paths(images_dir, masks_dir, num_samples)\n","\n"," for i, (img_path, mask_path) in enumerate(zip(image_paths, mask_paths)):\n"," # Charger avec PIL\n"," img = np.array(PILImage.open(img_path).resize((512, 256), PILImage.Resampling.BILINEAR))\n"," mask_raw = np.array(PILImage.open(mask_path).resize((512, 256), PILImage.Resampling.NEAREST))\n","\n"," # Appliquer le mapping (vérifier que id_to_group existe)\n"," if 'id_to_group' in globals():\n"," mask = id_to_group[mask_raw]\n"," else:\n"," print(\"⚠️ Variable 'id_to_group' non trouvée, utilisation brute\")\n"," mask = mask_raw\n","\n"," # Normaliser l'image pour l'affichage\n"," img_normalized = img.astype(np.float32) / 255.0\n","\n"," # Créer la visualisation\n"," fig = _create_sample_plot(img_normalized, mask, f\"{split.title()} Raw Sample {i+1}\")\n","\n"," # Sauvegarder si demandé\n"," if save_dir:\n"," os.makedirs(save_dir, exist_ok=True)\n"," fig_path = os.path.join(save_dir, f'{split}_raw_sample_{i+1}.png')\n"," fig.savefig(fig_path, dpi=150, bbox_inches='tight')\n"," print(f\"💾 Sauvegardé: {fig_path}\")\n","\n"," plt.show()\n"," plt.close(fig)\n","\n"," # Afficher les stats si demandé\n"," if show_stats:\n"," _show_class_stats(mask)\n","\n","def _create_sample_plot(image, mask, title):\n"," \"\"\"Crée le plot avec les 3 vues (factorisation du code de visualisation)\"\"\"\n","\n"," fig, axes = plt.subplots(1, 3, figsize=(15, 5))\n"," fig.suptitle(title, fontsize=14)\n","\n"," # Vue 1: Image originale\n"," axes[0].imshow(image)\n"," axes[0].set_title(\"Image Originale\")\n"," axes[0].axis('off')\n","\n"," # Vue 2: Masque avec IDs (en tab10)\n"," im = axes[1].imshow(mask, cmap='tab10', vmin=0, vmax=mapping_config['num_classes']-1)\n"," axes[1].set_title(\"Masque (IDs de classes)\")\n"," axes[1].axis('off')\n","\n"," # Vue 3: Masque colorisé avec nos couleurs\n"," colored_mask = _colorize_mask(mask)\n"," axes[2].imshow(colored_mask)\n"," axes[2].set_title(\"Masque Segmenté (Coloré)\")\n"," axes[2].axis('off')\n","\n"," plt.tight_layout()\n"," return fig\n","\n","def _colorize_mask(mask):\n"," \"\"\"Applique la colorisation du masque (logique centralisée)\"\"\"\n"," colored_mask = np.zeros((mask.shape[0], mask.shape[1], 3), dtype=np.uint8)\n"," for group_id in range(mapping_config['num_classes']):\n"," colored_mask[mask == group_id] = mapping_config['group_colors'][group_id]\n"," return colored_mask\n","\n","def _show_class_stats(mask):\n"," \"\"\"Affiche les statistiques des classes (factorisation)\"\"\"\n"," unique_classes, counts = np.unique(mask, return_counts=True)\n"," print(\"\\n📊 Classes présentes dans cette image:\")\n"," for class_id, count in zip(unique_classes, counts):\n"," if class_id < len(mapping_config['group_names']): # Vérification bounds\n"," percentage = (count / mask.size) * 100\n"," print(f\" {mapping_config['group_names'][class_id]}: {count} pixels ({percentage:.1f}%)\")\n","\n","def _collect_sample_paths(images_dir, masks_dir, num_samples):\n"," \"\"\"Collecte aléatoirement des chemins d'images (factorisation)\"\"\"\n"," image_paths, mask_paths = [], []\n","\n"," for city in os.listdir(images_dir):\n"," city_img_dir = os.path.join(images_dir, city)\n"," city_mask_dir = os.path.join(masks_dir, city)\n","\n"," if os.path.isdir(city_img_dir) and os.path.isdir(city_mask_dir):\n"," for file_name in os.listdir(city_img_dir):\n"," if file_name.endswith('_leftImg8bit.png'):\n"," img_path = os.path.join(city_img_dir, file_name)\n"," mask_name = file_name.replace('_leftImg8bit.png', '_gtFine_labelIds.png')\n"," mask_path = os.path.join(city_mask_dir, mask_name)\n","\n"," if os.path.exists(mask_path):\n"," image_paths.append(img_path)\n"," mask_paths.append(mask_path)\n","\n"," # Sélection aléatoire\n"," if len(image_paths) > num_samples:\n"," selected_indices = np.random.choice(len(image_paths), size=num_samples, replace=False)\n"," image_paths = [image_paths[i] for i in selected_indices]\n"," mask_paths = [mask_paths[i] for i in selected_indices]\n","\n"," return image_paths, mask_paths\n","\n","def _create_legend(save_dir):\n"," \"\"\"Crée la légende des classes (factorisation)\"\"\"\n"," fig, ax = plt.subplots(figsize=(8, 5))\n"," fig.suptitle(\"Légende des classes de segmentation\", fontsize=14)\n","\n"," for i, (name, color) in enumerate(zip(mapping_config['group_names'], mapping_config['group_colors'])):\n"," ax.fill_between([0, 1], [i, i], [i+0.9, i+0.9], color=[c/255 for c in color])\n"," ax.text(1.1, i+0.5, name, va='center', fontsize=12)\n","\n"," ax.set_xlim(0, 2.5)\n"," ax.set_ylim(len(mapping_config['group_names'])-0.1, -0.1)\n"," ax.axis('off')\n","\n"," if save_dir:\n"," os.makedirs(save_dir, exist_ok=True)\n"," legend_path = os.path.join(save_dir, 'class_legend.png')\n"," fig.savefig(legend_path, dpi=150, bbox_inches='tight')\n"," print(f\"🏷️ Légende sauvegardée: {legend_path}\")\n","\n"," plt.show()\n"," plt.close(fig)\n","\n","# === UTILISATION SIMPLIFIÉE ===\n","print(\"=== VISUALISATION DES ÉCHANTILLONS CITYSCAPES ===\")\n","\n","# Méthode préférée: depuis les datasets TensorFlow\n","datasets_dict = {\n"," 'train': train_dataset,\n"," 'val': val_dataset,\n"," 'test': test_dataset\n","}\n","\n","# Créer le dossier de sauvegarde\n","viz_dir = os.path.join(data_dir, 'samples') if 'data_dir' in globals() else './visualizations'\n","\n","# Visualiser avec toutes les options\n","visualize_cityscapes_samples(\n"," datasets_dict=datasets_dict,\n"," num_samples=2, # 2 échantillons par dataset\n"," save_dir=viz_dir,\n"," show_stats=True,\n"," show_legend=True\n",")"]},{"cell_type":"markdown","metadata":{"id":"BcRXJIz6Ts6q"},"source":["## Entrainement du modèle"]},{"cell_type":"code","execution_count":10,"metadata":{"colab":{"base_uri":"https://localhost:8080/"},"executionInfo":{"elapsed":4,"status":"ok","timestamp":1755008364624,"user":{"displayName":"Grégoire Mureau","userId":"04855910922109670788"},"user_tz":-120},"id":"lUxAq504vmjH","outputId":"8b3d5f4d-a841-4a14-bb1c-8af33d056fe6"},"outputs":[{"output_type":"stream","name":"stdout","text":["🚀 MODE PRODUCTION ACTIVÉ\n","📊 Configuration globale: Device=/CPU:0, Max_Epochs=20\n"," Patience EarlyStopping=5, Patience LR=3\n"]}],"source":["# ====================================================================\n","# SCRIPT D'ENTRAÎNEMENT OPTIMISÉ POUR SEGMENTATION MULTI-CONFIGURATIONS\n","# AVEC SYSTÈME DE REPRISE AUTOMATIQUE\n","# ====================================================================\n","\n","# 🚨 PARAMÈTRE PRINCIPAL : Basculer entre TEST et PRODUCTION\n","TEST_MODE = False # Mettre False pour le mode production\n","\n","# Configuration selon le mode\n","if TEST_MODE:\n"," print(\"🧪 MODE TEST ACTIVÉ\")\n"," DEVICE_CONFIG = '/CPU:0'\n"," MAX_EPOCHS = 2 # Très rapide pour les tests\n"," EARLY_STOPPING_PATIENCE = 2 # Arrêt anticipé si pas d'amélioration pendant 2 époques\n"," LR_REDUCE_PATIENCE = 1 # Réduction LR si pas d'amélioration pendant 1 époque\n"," SUBSET_SIZE = 20\n"," # Taille fixe pour les tests (plus rapide)\n"," TEST_IMG_SIZE = (128, 128)\n"," TEST_BATCH_SIZE = 2\n"," TEST_MAX_CONFIGS = None # Nombre max de configs à tester\n"," # Expérience MLflow déjà définie en début de notebook\n","else:\n"," print(\"🚀 MODE PRODUCTION ACTIVÉ\")\n"," DEVICE_CONFIG = '/GPU:0' if tf.config.list_physical_devices('GPU') else '/CPU:0'\n"," MAX_EPOCHS = 20 # 20 époques max (beaucoup de modèles à tester)\n"," EARLY_STOPPING_PATIENCE = 5 # Arrêt anticipé si pas d'amélioration pendant 5 époques\n"," LR_REDUCE_PATIENCE = 3 # Réduction LR si pas d'amélioration pendant 3 époques\n"," SUBSET_SIZE = None\n"," # En production, utilise les paramètres de chaque configuration\n"," # Expérience MLflow déjà définie en début de notebook\n","\n","print(f\"📊 Configuration globale: Device={DEVICE_CONFIG}, Max_Epochs={MAX_EPOCHS}\")\n","print(f\" Patience EarlyStopping={EARLY_STOPPING_PATIENCE}, Patience LR={LR_REDUCE_PATIENCE}\")"]},{"cell_type":"code","execution_count":11,"metadata":{"id":"RqI-umcJvt9u","executionInfo":{"status":"ok","timestamp":1755008364680,"user_tz":-120,"elapsed":55,"user":{"displayName":"Grégoire Mureau","userId":"04855910922109670788"}}},"outputs":[],"source":["# ====================================================================\n","# SYSTÈME DE REPRISE AUTOMATIQUE - VÉRIFICATION MLFLOW\n","# ====================================================================\n","\n","def get_completed_experiments(experiment_name=\"OC Projet 8\"):\n"," \"\"\"\n"," Récupère la liste des expériences déjà terminées dans MLflow\n"," Retourne un set des noms de runs terminés\n"," \"\"\"\n"," try:\n"," print(f\"🔍 Vérification des expériences déjà terminées dans '{experiment_name}'...\")\n","\n"," # Récupérer l'expérience par nom\n"," experiment = mlflow.get_experiment_by_name(experiment_name)\n"," if experiment is None:\n"," print(f\"⚠️ Expérience '{experiment_name}' non trouvée\")\n"," return set()\n","\n"," # Rechercher tous les runs de cette expérience\n"," runs = mlflow.search_runs(\n"," experiment_ids=[experiment.experiment_id],\n"," filter_string=\"status = 'FINISHED'\", # Seulement les runs terminés avec succès\n"," output_format=\"pandas\"\n"," )\n","\n"," if runs.empty:\n"," print(\"📊 Aucune expérience terminée trouvée\")\n"," return set()\n","\n"," # Extraire les noms des runs terminés\n"," completed_runs = set()\n"," for idx, run in runs.iterrows():\n"," run_name = run.get('tags.mlflow.runName', '')\n"," if run_name:\n"," completed_runs.add(run_name)\n","\n"," print(f\"✅ {len(completed_runs)} expériences terminées trouvées:\")\n"," for run_name in sorted(completed_runs):\n"," print(f\" - {run_name}\")\n","\n"," return completed_runs\n","\n"," except Exception as e:\n"," print(f\"⚠️ Erreur lors de la vérification MLflow: {e}\")\n"," print(\"🔄 Continuera sans vérification (tous les modèles seront entraînés)\")\n"," return set()\n","\n","\n","def generate_stable_run_name(config, test_mode=False):\n"," \"\"\"\n"," Génère un nom de run stable (sans date) basé sur la configuration\n"," Format: {config_name}_{mode}\n"," \"\"\"\n"," base_name = config['name']\n"," mode_suffix = \"test\" if test_mode else \"prod\"\n"," return f\"{base_name}_{mode_suffix}\"\n","\n","\n","def filter_configs_to_run(experimental_configurations, experiment_name=\"OC Projet 8\"):\n"," \"\"\"\n"," Filtre les configurations pour ne garder que celles non encore entraînées\n"," \"\"\"\n"," completed_runs = get_completed_experiments(experiment_name)\n","\n"," if not completed_runs:\n"," print(\"🚀 Aucune expérience terminée trouvée, tous les modèles seront entraînés\")\n"," return list(range(len(experimental_configurations)))\n","\n"," configs_to_run = []\n"," skipped_configs = []\n","\n"," for idx, config in enumerate(experimental_configurations):\n"," expected_run_name = generate_stable_run_name(config, TEST_MODE)\n","\n"," if expected_run_name in completed_runs:\n"," skipped_configs.append((idx, config['name'], expected_run_name))\n"," else:\n"," configs_to_run.append(idx)\n","\n"," # Rapport de filtrage\n"," print(f\"\\n📊 FILTRAGE DES CONFIGURATIONS:\")\n"," print(f\"✅ À entraîner: {len(configs_to_run)} configurations\")\n"," print(f\"⏭️ Déjà terminées: {len(skipped_configs)} configurations\")\n","\n"," if skipped_configs:\n"," print(f\"\\n⏭️ CONFIGURATIONS IGNORÉES (déjà terminées):\")\n"," for idx, config_name, run_name in skipped_configs:\n"," print(f\" [{idx}] {config_name} → {run_name}\")\n","\n"," if configs_to_run:\n"," print(f\"\\n🚀 CONFIGURATIONS À ENTRAÎNER:\")\n"," for idx in configs_to_run:\n"," config_name = experimental_configurations[idx]['name']\n"," expected_run_name = generate_stable_run_name(experimental_configurations[idx], TEST_MODE)\n"," print(f\" [{idx}] {config_name} → {expected_run_name}\")\n","\n"," return configs_to_run"]},{"cell_type":"code","execution_count":12,"metadata":{"id":"yXkQoZssvwTt","executionInfo":{"status":"ok","timestamp":1755008364708,"user_tz":-120,"elapsed":27,"user":{"displayName":"Grégoire Mureau","userId":"04855910922109670788"}}},"outputs":[],"source":["# ====================================================================\n","# CALLBACK MLFLOW PERSONNALISÉ POUR LOGGING DÉTAILLÉ\n","# ====================================================================\n","\n","class ExperimentMLflowCallback(tf.keras.callbacks.Callback):\n"," \"\"\"Callback MLflow personnalisé pour logging complet des métriques\"\"\"\n","\n"," def __init__(self, config, run_id, batch_size, img_size):\n"," super().__init__()\n"," self.config = config\n"," self.run_id = run_id\n"," self.batch_size = batch_size\n"," self.img_size = img_size\n","\n"," def on_epoch_end(self, epoch, logs=None):\n"," \"\"\"Log des métriques à chaque époque (sans doublons)\"\"\"\n"," logs = logs or {}\n"," for name, value in logs.items():\n"," # Éviter les doublons : ignorer les métriques \"validation_\"\n"," if not name.startswith('validation_'):\n"," mlflow.log_metric(name, float(value), step=epoch)\n","\n"," def on_train_begin(self, logs=None):\n"," \"\"\"Log des paramètres au début de l'entraînement\"\"\"\n"," # Log config du modèle\n"," mlflow.log_param(\"model_architecture\", self.config[\"architecture\"])\n"," mlflow.log_param(\"model_name\", self.config[\"name\"])\n"," mlflow.log_param(\"encoder\", self.config.get(\"encoder\", \"None\"))\n"," mlflow.log_param(\"encoder_freeze\", self.config.get(\"encoder_freeze\", False))\n","\n"," # Log config du script (utilise les paramètres passés au callback)\n"," mlflow.log_param(\"script_batch_size\", self.batch_size)\n"," mlflow.log_param(\"script_img_size\", str(self.img_size))\n"," mlflow.log_param(\"test_mode\", TEST_MODE)\n"," mlflow.log_param(\"max_epochs\", MAX_EPOCHS)\n"," mlflow.log_param(\"device\", DEVICE_CONFIG)\n"," mlflow.log_param(\"subset_size\", SUBSET_SIZE)\n","\n","\n","class KerasModelCheckpoint(tf.keras.callbacks.Callback):\n"," \"\"\"Callback personnalisé pour sauvegarder en format .keras natif\"\"\"\n","\n"," def __init__(self, filepath, monitor='val_loss', save_best_only=True, mode='auto', verbose=1):\n"," super().__init__()\n"," self.filepath = filepath\n"," self.monitor = monitor\n"," self.save_best_only = save_best_only\n"," self.verbose = verbose\n","\n"," if mode == 'auto':\n"," if 'acc' in monitor or monitor.startswith('fmeasure'):\n"," self.mode = 'max'\n"," else:\n"," self.mode = 'min'\n"," else:\n"," self.mode = mode\n","\n"," if self.mode == 'min':\n"," self.best = float('inf')\n"," else:\n"," self.best = -float('inf')\n","\n"," def on_epoch_end(self, epoch, logs=None):\n"," logs = logs or {}\n"," current = logs.get(self.monitor)\n","\n"," if current is None:\n"," if self.verbose > 0:\n"," print(f'Warning: Metric {self.monitor} not found in logs.')\n"," return\n","\n"," if self.save_best_only:\n"," if (self.mode == 'min' and current < self.best) or (self.mode == 'max' and current > self.best):\n"," self.best = current\n","\n"," if self.verbose > 0:\n"," print(f'Epoch {epoch + 1}: {self.monitor} improved from {self.best} to {current:.5f}, saving model to {self.filepath}')\n","\n"," # SAUVEGARDE NATIVE KERAS - AUCUN WARNING !\n"," self.model.save(self.filepath, save_format='keras')\n"," else:\n"," if self.verbose > 0:\n"," print(f'Epoch {epoch + 1}: {self.monitor} did not improve from {self.best:.5f}')\n"," else:\n"," if self.verbose > 0:\n"," print(f'Epoch {epoch + 1}: saving model to {self.filepath}')\n"," self.model.save(self.filepath, save_format='keras')"]},{"cell_type":"code","source":["def calculate_detailed_metrics(model, test_dataset, num_classes, class_names=None):\n"," \"\"\"\n"," Calcule uniquement l'IoU par classe et les statistiques détaillées\n"," (Le mean_iou global est maintenant calculé par la métrique officielle)\n"," \"\"\"\n"," print(\" 📊 Analyse détaillée par classe...\")\n","\n"," all_true_masks = []\n"," all_pred_masks = []\n","\n"," # Collecter toutes les prédictions\n"," for batch_idx, (images, true_masks) in enumerate(test_dataset):\n"," predictions = model.predict(images, verbose=0)\n"," pred_masks = tf.argmax(predictions, axis=-1)\n","\n"," all_true_masks.extend(true_masks.numpy())\n"," all_pred_masks.extend(pred_masks.numpy())\n","\n"," if batch_idx % 10 == 0:\n"," print(f\" Batch {batch_idx + 1} traité...\")\n","\n"," all_true_masks = np.array(all_true_masks)\n"," all_pred_masks = np.array(all_pred_masks)\n","\n"," print(f\" Total d'images évaluées: {len(all_true_masks)}\")\n","\n"," # Calcul IoU par classe (pour analyse détaillée)\n"," class_ious = []\n"," class_stats = []\n","\n"," for class_id in range(num_classes):\n"," # Masques binaires pour cette classe\n"," true_binary = (all_true_masks == class_id)\n"," pred_binary = (all_pred_masks == class_id)\n","\n"," # Calcul IoU\n"," intersection = np.logical_and(true_binary, pred_binary).sum()\n"," union = np.logical_or(true_binary, pred_binary).sum()\n","\n"," iou = intersection / union if union > 0 else 0.0\n"," class_ious.append(iou)\n","\n"," # Statistiques de classe\n"," true_pixels = true_binary.sum()\n"," pred_pixels = pred_binary.sum()\n"," total_pixels = all_true_masks.size\n","\n"," class_name = class_names[class_id] if class_names else f\"Class_{class_id}\"\n","\n"," class_stats.append({\n"," 'class_id': class_id,\n"," 'class_name': class_name,\n"," 'iou': iou,\n"," 'true_pixels': true_pixels,\n"," 'pred_pixels': pred_pixels,\n"," 'true_percentage': 100 * true_pixels / total_pixels,\n"," 'pred_percentage': 100 * pred_pixels / total_pixels\n"," })\n","\n"," # Mean IoU calculé ici (pour cohérence avec l'affichage)\n"," mean_iou = np.mean(class_ious)\n","\n"," # Affichage des résultats détaillés\n"," print(f\"\\n📊 ANALYSE DÉTAILLÉE PAR CLASSE:\")\n"," print(\"-\" * 70)\n"," print(f\"{'Classe':<15} {'IoU':<8} {'Vrais %':<10} {'Prédits %':<12}\")\n"," print(\"-\" * 70)\n","\n"," for stats in class_stats:\n"," print(f\"{stats['class_name']:<15} {stats['iou']:<8.4f} {stats['true_percentage']:<10.2f} {stats['pred_percentage']:<12.2f}\")\n","\n"," print(\"-\" * 70)\n"," print(f\"{'MEAN IoU (calc)':<15} {mean_iou:<8.4f}\")\n"," print(\"💡 Le Mean IoU officiel (métrique Keras) fait référence\")\n","\n"," return {\n"," 'class_ious': class_ious,\n"," 'mean_iou': mean_iou, # Pour compatibilité avec le code existant\n"," 'class_stats': class_stats,\n"," 'all_true_masks': all_true_masks,\n"," 'all_pred_masks': all_pred_masks\n"," }\n"],"metadata":{"id":"9-rX20362Pz2","executionInfo":{"status":"ok","timestamp":1755008364728,"user_tz":-120,"elapsed":17,"user":{"displayName":"Grégoire Mureau","userId":"04855910922109670788"}}},"execution_count":13,"outputs":[]},{"cell_type":"code","source":["def setup_training_components(config, run_id):\n"," \"\"\"\n"," Setup complet : paramètres, datasets, modèle, compilation, callbacks\n"," VERSION CORRIGÉE avec métriques officielles Keras\n"," \"\"\"\n"," # === EXTRACTION DES PARAMÈTRES DE LA CONFIG ===\n"," if TEST_MODE:\n"," batch_size = TEST_BATCH_SIZE\n"," img_size = TEST_IMG_SIZE\n"," print(f\"🧪 Mode test: batch_size={batch_size}, img_size={img_size}\")\n"," else:\n"," batch_size = config['params']['batch_size']\n"," img_size = config['params']['img_size']\n"," print(f\"🚀 Mode production: batch_size={batch_size}, img_size={img_size}\")\n","\n"," # === PRÉPARATION DES DONNÉES ===\n"," print(\"\\n📊 Préparation des datasets...\")\n"," try:\n"," id_mapping = id_to_group\n"," mlflow.log_param(\"class_mapping\", \"custom_id_to_group\")\n"," except NameError:\n"," print(\"⚠️ Variable 'id_to_group' non définie\")\n"," id_mapping = None\n"," mlflow.log_param(\"class_mapping\", \"none\")\n","\n"," # Créer les datasets avec les bons paramètres\n"," train_dataset, val_dataset = prepare_cityscapes_dataset(\n"," cityscapes_dir,\n"," split='train',\n"," batch_size=batch_size,\n"," img_size=img_size,\n"," validation_split=0.2,\n"," id_to_group_mapping=id_mapping\n"," )\n"," test_dataset = prepare_cityscapes_dataset(\n"," cityscapes_dir,\n"," split='val',\n"," batch_size=batch_size,\n"," img_size=img_size,\n"," id_to_group_mapping=id_mapping\n"," )\n","\n"," if TEST_MODE and SUBSET_SIZE:\n"," train_batches = SUBSET_SIZE // batch_size\n"," val_batches = SUBSET_SIZE // (batch_size * 4)\n"," test_batches = SUBSET_SIZE // (batch_size * 4)\n"," train_samples_used = train_batches * batch_size\n"," val_samples_used = val_batches * batch_size\n"," test_samples_used = test_batches * batch_size\n","\n"," print(f\"🧪 Mode test: limitation appliquée\")\n"," print(f\" └─ 📚 Train utilisé: {train_samples_used} images ({train_batches} batches)\")\n"," print(f\" └─ 📊 Validation utilisé: {val_samples_used} images ({val_batches} batches)\")\n"," print(f\" └─ 🧪 Test utilisé: {test_samples_used} images ({test_batches} batches)\")\n","\n"," train_dataset = train_dataset.take(train_batches)\n"," val_dataset = val_dataset.take(val_batches)\n"," test_dataset = test_dataset.take(test_batches)\n","\n"," # === CRÉATION DU MODÈLE ===\n"," print(f\"\\n🏗️ Création du modèle {config['name']}...\")\n"," if TEST_MODE:\n"," config_modified = config.copy()\n"," config_modified['params'] = config['params'].copy()\n"," config_modified['params']['img_size'] = img_size\n"," config_modified['params']['batch_size'] = batch_size\n"," model_config = config_modified\n"," else:\n"," model_config = config\n","\n"," with tf.device(DEVICE_CONFIG):\n"," model = create_model_from_config(model_config, mapping_config['num_classes'])\n"," num_classes_from_model = model.output_shape[-1]\n"," print(f\"📊 Nombre de classes: {num_classes_from_model}\")\n"," total_params = model.count_params()\n"," trainable_params = sum([tf.keras.backend.count_params(w) for w in model.trainable_weights])\n"," print(f\"📊 Paramètres: {total_params:,} total | {trainable_params:,} entraînables\")\n","\n"," # === COMPILATION AVEC MÉTRIQUES OFFICIELLES KERAS ===\n"," print(\"\\n⚙️ Compilation du modèle avec métriques officielles...\")\n","\n"," # ✅ MÉTRIQUES OFFICIELLES KERAS - AUCUNE MÉTRIQUE PERSONNALISÉE !\n"," metrics_list = [\n"," # Accuracy officielle pour classification sparse\n"," tf.keras.metrics.SparseCategoricalAccuracy(name='sparse_categorical_accuracy'),\n","\n"," # IoU officielle Keras - LA métrique de référence pour segmentation\n"," tf.keras.metrics.MeanIoU(\n"," num_classes=num_classes_from_model,\n"," name='mean_iou',\n"," sparse_y_true=True, # Les labels sont des entiers (pas one-hot)\n"," sparse_y_pred=False # Les prédictions sont des probabilités\n"," )\n"," ]\n","\n"," # Métrique principale pour les callbacks\n"," primary_metric = 'val_mean_iou'\n"," primary_mode = 'max'\n","\n"," print(\"🎯 MÉTRIQUES OFFICIELLES KERAS POUR SEGMENTATION SÉMANTIQUE:\")\n"," print(\" 🔥 LOSS: sparse_categorical_crossentropy\")\n"," print(\" └─ Optimise la classification pixel-par-pixel\")\n"," print(\" 📊 MÉTRIQUES (100% officielles Keras):\")\n"," print(\" ├─ SparseCategoricalAccuracy: % pixels correctement classifiés\")\n"," print(\" └─ MeanIoU: Intersection over Union (métrique gold standard)\")\n"," print(\" 🎛️ CALLBACKS:\")\n"," print(f\" └─ Monitor: {primary_metric} (IoU = métrique de référence)\")\n"," print(\"✅ Configuration 100% standard et fiable !\")\n","\n"," # Compilation du modèle\n"," model.compile(\n"," optimizer=tf.keras.optimizers.Adam(learning_rate=0.001),\n"," loss='sparse_categorical_crossentropy',\n"," metrics=metrics_list\n"," )\n","\n"," # === CALLBACKS OPTIMISÉS ===\n"," print(f\"\\n📋 Configuration des callbacks...\")\n"," print(\"🎛️ CALLBACKS BASÉS SUR MEAN_IOU OFFICIEL:\")\n"," print(f\" ├─ EarlyStopping: Arrêt si {primary_metric} stagne {EARLY_STOPPING_PATIENCE} époques\")\n"," print(f\" ├─ ReduceLROnPlateau: Réduction LR si {primary_metric} stagne {LR_REDUCE_PATIENCE} époques\")\n"," print(f\" └─ ModelCheckpoint: Sauvegarde du meilleur modèle selon {primary_metric}\")\n","\n"," artifacts_dir = f\"artifacts/run_{run_id}\"\n"," os.makedirs(artifacts_dir, exist_ok=True)\n","\n"," mlflow_callback = ExperimentMLflowCallback(config, run_id, batch_size, img_size)\n","\n"," callbacks_list = [\n"," mlflow_callback,\n"," tf.keras.callbacks.ReduceLROnPlateau(\n"," monitor=primary_metric,\n"," factor=0.2,\n"," patience=LR_REDUCE_PATIENCE,\n"," verbose=1,\n"," min_lr=1e-6,\n"," mode=primary_mode\n"," ),\n"," tf.keras.callbacks.EarlyStopping(\n"," monitor=primary_metric,\n"," patience=EARLY_STOPPING_PATIENCE,\n"," restore_best_weights=True,\n"," verbose=1,\n"," mode=primary_mode\n"," ),\n"," KerasModelCheckpoint(\n"," filepath=os.path.join(artifacts_dir, 'best_model.keras'),\n"," monitor=primary_metric,\n"," save_best_only=True,\n"," mode=primary_mode,\n"," verbose=1\n"," ),\n"," tf.keras.callbacks.TensorBoard(\n"," log_dir=os.path.join(artifacts_dir, 'tensorboard'),\n"," histogram_freq=1\n"," )\n"," ]\n","\n"," return {\n"," 'model': model,\n"," 'train_dataset': train_dataset,\n"," 'val_dataset': val_dataset,\n"," 'test_dataset': test_dataset,\n"," 'callbacks_list': callbacks_list,\n"," 'artifacts_dir': artifacts_dir,\n"," 'num_classes_from_model': num_classes_from_model,\n"," 'batch_size': batch_size,\n"," 'img_size': img_size\n"," }\n","\n","def train_model_core(model, train_dataset, val_dataset, callbacks_list, max_epochs):\n"," \"\"\"\n"," Fonction pure d'entraînement - ne fait que l'entraînement\n"," \"\"\"\n"," print(f\"\\n🚀 Début de l'entraînement ({max_epochs} epochs max)...\")\n"," start_time = time.time()\n","\n"," with tf.device(DEVICE_CONFIG):\n"," history = model.fit(\n"," train_dataset,\n"," validation_data=val_dataset,\n"," epochs=max_epochs,\n"," callbacks=callbacks_list,\n"," verbose=1\n"," )\n","\n"," training_time = time.time() - start_time\n"," mlflow.log_metric(\"training_time_seconds\", training_time)\n"," print(f\"⏱️ Temps d'entraînement: {training_time:.2f}s\")\n","\n"," return history, training_time\n","\n","def evaluate_model_comprehensive(model, test_dataset, num_classes_from_model, artifacts_dir):\n"," \"\"\"\n"," Évaluation complète avec métriques officielles Keras + analyses détaillées\n"," \"\"\"\n"," # === ÉVALUATION AVEC MÉTRIQUES OFFICIELLES (même config que l'entraînement) ===\n"," print(\"\\n📊 Évaluation avec métriques officielles Keras...\")\n","\n"," # Créer exactement les mêmes métriques que pendant l'entraînement\n"," official_metrics = [\n"," tf.keras.metrics.SparseCategoricalAccuracy(name='sparse_categorical_accuracy'),\n"," tf.keras.metrics.MeanIoU(\n"," num_classes=num_classes_from_model,\n"," name='mean_iou',\n"," sparse_y_true=True,\n"," sparse_y_pred=False\n"," )\n"," ]\n","\n"," # Compiler temporairement avec les métriques officielles\n"," # (ça ne change pas les poids, juste les métriques)\n"," model.compile(\n"," optimizer=model.optimizer, # Garder l'optimizer existant\n"," loss=model.loss, # Garder la loss existante\n"," metrics=official_metrics\n"," )\n","\n"," # Évaluation officielle\n"," test_results = model.evaluate(test_dataset, verbose=1)\n","\n"," basic_test_metrics = {\n"," 'loss': test_results[0],\n"," 'sparse_categorical_accuracy': test_results[1],\n"," 'mean_iou': test_results[2]\n"," }\n","\n"," # Log dans MLflow\n"," for metric_name, value in basic_test_metrics.items():\n"," mlflow.log_metric(f\"test_{metric_name}\", value)\n"," print(f\"✅ Test {metric_name} (officiel): {value:.4f}\")\n","\n"," # === ÉVALUATION DÉTAILLÉE ADDITIONNELLE (pour analyse par classe) ===\n"," try:\n"," class_names = mapping_config['group_names'] if 'group_names' in mapping_config else None\n"," except:\n"," class_names = [f\"Class_{i}\" for i in range(num_classes_from_model)]\n","\n"," print(\"\\n🔍 Calcul des métriques détaillées par classe...\")\n"," detailed_metrics = calculate_detailed_metrics(\n"," model, test_dataset, num_classes_from_model, class_names\n"," )\n","\n"," # Log des métriques par classe (en complément)\n"," for i, iou in enumerate(detailed_metrics['class_ious']):\n"," mlflow.log_metric(f\"class_{i}_iou\", iou)\n","\n"," # === MATRICE DE CONFUSION ===\n"," print(\"\\n🔥 Génération de la matrice de confusion...\")\n"," confusion_matrix_data = generate_confusion_matrix(\n"," model, test_dataset, num_classes_from_model, class_names, artifacts_dir\n"," )\n","\n"," return basic_test_metrics, detailed_metrics, confusion_matrix_data, class_names\n","\n","def generate_confusion_matrix(model, test_dataset, num_classes, class_names, artifacts_dir):\n"," \"\"\"\n"," Génère la matrice de confusion sur le jeu de test et la sauvegarde (SANS AFFICHAGE)\n"," \"\"\"\n"," print(\"📊 Calcul de la matrice de confusion...\")\n","\n"," # Collecter toutes les prédictions\n"," y_true_all = []\n"," y_pred_all = []\n","\n"," for batch_images, batch_masks in test_dataset:\n"," # Prédictions du modèle\n"," predictions = model.predict(batch_images, verbose=0)\n"," pred_classes = tf.argmax(predictions, axis=-1)\n","\n"," # Aplatir les masques et prédictions\n"," true_flat = tf.reshape(batch_masks, [-1]).numpy()\n"," pred_flat = tf.reshape(pred_classes, [-1]).numpy()\n","\n"," y_true_all.extend(true_flat)\n"," y_pred_all.extend(pred_flat)\n","\n"," # Calcul de la matrice de confusion\n"," from sklearn.metrics import confusion_matrix\n"," import itertools\n","\n"," cm = confusion_matrix(y_true_all, y_pred_all, labels=range(num_classes))\n","\n"," # === CRÉATION DE LA VISUALISATION (SANS AFFICHAGE) ===\n"," plt.figure(figsize=(10, 8))\n"," plt.imshow(cm, interpolation='nearest', cmap=plt.cm.Blues)\n"," plt.title('Matrice de Confusion - Jeu de Test', fontsize=14)\n"," plt.colorbar()\n","\n"," # Définir les labels des axes\n"," if class_names:\n"," tick_marks = np.arange(len(class_names))\n"," plt.xticks(tick_marks, class_names, rotation=45, ha='right')\n"," plt.yticks(tick_marks, class_names)\n"," else:\n"," tick_marks = np.arange(num_classes)\n"," plt.xticks(tick_marks, [f\"Class_{i}\" for i in range(num_classes)])\n"," plt.yticks(tick_marks, [f\"Class_{i}\" for i in range(num_classes)])\n","\n"," # === AFFICHER LES VALEURS DANS CHAQUE CELLULE ===\n"," thresh = cm.max() / 2.0\n"," for i, j in itertools.product(range(cm.shape[0]), range(cm.shape[1])):\n"," plt.text(j, i, format(cm[i, j], 'd'),\n"," horizontalalignment=\"center\",\n"," verticalalignment=\"center\",\n"," color=\"white\" if cm[i, j] > thresh else \"black\",\n"," fontsize=10,\n"," fontweight='bold')\n","\n"," plt.ylabel('Vraie classe', fontsize=12)\n"," plt.xlabel('Classe prédite', fontsize=12)\n"," plt.tight_layout()\n","\n"," # Sauvegarder l'image (SANS AFFICHAGE)\n"," cm_image_path = os.path.join(artifacts_dir, 'confusion_matrix.png')\n"," plt.savefig(cm_image_path, dpi=300, bbox_inches='tight')\n"," plt.close() # ✅ FERMER LE PLOT - PAS D'AFFICHAGE\n","\n"," # Données JSON pour l'artefact\n"," confusion_matrix_data = {\n"," 'matrix': cm.tolist(), # Conversion numpy -> liste Python\n"," 'class_names': class_names if class_names else [f\"Class_{i}\" for i in range(num_classes)],\n"," 'total_samples': len(y_true_all),\n"," 'accuracy': float(np.trace(cm) / np.sum(cm)) # Accuracy globale\n"," }\n","\n"," print(f\"✅ Matrice de confusion sauvegardée : {cm_image_path}\")\n","\n"," return confusion_matrix_data\n","\n","def generate_training_visualizations(history, detailed_metrics, class_names, run_id, config, artifacts_dir):\n"," \"\"\"\n"," Génère toutes les visualisations : courbes d'entraînement + IoU par classe\n"," \"\"\"\n"," print(\"\\n📈 Création des visualisations...\")\n"," fig, axes = plt.subplots(2, 2, figsize=(15, 10))\n"," fig.suptitle(f'Entraînement {config[\"name\"]} - Run {run_id[:8]}', fontsize=16)\n","\n"," # Loss\n"," axes[0, 0].plot(history.history['loss'], label='Train Loss')\n"," axes[0, 0].plot(history.history['val_loss'], label='Val Loss')\n"," axes[0, 0].set_title('Loss')\n"," axes[0, 0].legend()\n"," axes[0, 0].grid(True)\n","\n"," # Accuracy\n"," axes[0, 1].plot(history.history['sparse_categorical_accuracy'], label='Train Acc')\n"," axes[0, 1].plot(history.history['val_sparse_categorical_accuracy'], label='Val Acc')\n"," axes[0, 1].set_title('Accuracy')\n"," axes[0, 1].legend()\n"," axes[0, 1].grid(True)\n","\n"," # Mean IoU\n"," axes[1, 0].plot(history.history['mean_iou'], label='Train Mean IoU')\n"," axes[1, 0].plot(history.history['val_mean_iou'], label='Val Mean IoU')\n"," axes[1, 0].set_title('Mean IoU')\n"," axes[1, 0].legend()\n"," axes[1, 0].grid(True)\n","\n"," # IoU par classe (bar chart)\n"," if class_names and len(class_names) == len(detailed_metrics['class_ious']):\n"," axes[1, 1].bar(range(len(detailed_metrics['class_ious'])), detailed_metrics['class_ious'])\n"," axes[1, 1].set_title(f'IoU par classe sur jeu de test (Mean: {detailed_metrics[\"mean_iou\"]:.3f})')\n"," axes[1, 1].set_xticks(range(len(class_names)))\n"," axes[1, 1].set_xticklabels(class_names, rotation=45)\n"," axes[1, 1].grid(True, axis='y')\n","\n"," plt.tight_layout()\n"," training_curves_path = os.path.join(artifacts_dir, 'training_curves.png')\n"," plt.savefig(training_curves_path, dpi=300, bbox_inches='tight')\n"," plt.close()\n","\n","def save_training_artifacts(history, config, detailed_metrics, confusion_matrix_data, artifacts_dir):\n"," \"\"\"\n"," Sauvegarde tous les artefacts : history, config, métriques, matrice de confusion\n"," \"\"\"\n"," print(\"\\n💾 Sauvegarde des artefacts...\")\n","\n"," # Histoire d'entraînement\n"," history_path = os.path.join(artifacts_dir, 'training_history.json')\n"," with open(history_path, 'w') as f:\n"," history_json = {k: [float(v) for v in values] for k, values in history.history.items()}\n"," json.dump(history_json, f, indent=2)\n","\n"," # Configuration du modèle\n"," config_path = os.path.join(artifacts_dir, 'model_config.json')\n"," with open(config_path, 'w') as f:\n"," json.dump(config, f, indent=2)\n","\n"," # Métriques détaillées - CONVERSION NUMPY VERS PYTHON POUR JSON\n"," detailed_path = os.path.join(artifacts_dir, 'detailed_metrics.json')\n"," with open(detailed_path, 'w') as f:\n"," class_stats_json = []\n"," for stats in detailed_metrics['class_stats']:\n"," stats_converted = {\n"," 'class_id': int(stats['class_id']),\n"," 'class_name': str(stats['class_name']),\n"," 'iou': float(stats['iou']),\n"," 'true_pixels': int(stats['true_pixels']),\n"," 'pred_pixels': int(stats['pred_pixels']),\n"," 'true_percentage': float(stats['true_percentage']),\n"," 'pred_percentage': float(stats['pred_percentage'])\n"," }\n"," class_stats_json.append(stats_converted)\n","\n"," detailed_json = {\n"," 'mean_iou': float(detailed_metrics['mean_iou']),\n"," 'class_ious': [float(iou) for iou in detailed_metrics['class_ious']],\n"," 'class_stats': class_stats_json\n"," }\n"," json.dump(detailed_json, f, indent=2)\n","\n"," # Matrice de confusion (si disponible)\n"," if confusion_matrix_data:\n"," confusion_path = os.path.join(artifacts_dir, 'confusion_matrix.json')\n"," with open(confusion_path, 'w') as f:\n"," json.dump(confusion_matrix_data, f, indent=2)\n","\n"," mlflow.log_artifacts(artifacts_dir)\n","\n","def train_segmentation_model_optimized(config, run_name=None):\n"," \"\"\"\n"," Fonction principale orchestrant l'entraînement complet - VERSION REFACTORISÉE\n"," \"\"\"\n"," with mlflow.start_run(run_name=run_name) as run:\n"," run_id = run.info.run_id\n"," print(f\"🔬 MLflow Run ID: {run_id}\")\n"," print(f\"🏷️ MLflow Run Name: {run_name}\")\n","\n"," try:\n"," # === 1. SETUP COMPLET ===\n"," setup_components = setup_training_components(config, run_id)\n","\n"," # === 2. ENTRAÎNEMENT ===\n"," history, training_time = train_model_core(\n"," setup_components['model'],\n"," setup_components['train_dataset'],\n"," setup_components['val_dataset'],\n"," setup_components['callbacks_list'],\n"," MAX_EPOCHS\n"," )\n","\n"," # === 3. ÉVALUATION COMPLÈTE ===\n"," basic_test_metrics, detailed_metrics, confusion_matrix_data, class_names = evaluate_model_comprehensive(\n"," setup_components['model'],\n"," setup_components['test_dataset'],\n"," setup_components['num_classes_from_model'],\n"," setup_components['artifacts_dir']\n"," )\n","\n"," # === 4. VISUALISATIONS ===\n"," generate_training_visualizations(\n"," history, detailed_metrics, class_names,\n"," run_id, config, setup_components['artifacts_dir']\n"," )\n","\n"," # === 5. SAUVEGARDE DES ARTEFACTS ===\n"," save_training_artifacts(\n"," history, config, detailed_metrics,\n"," confusion_matrix_data, setup_components['artifacts_dir']\n"," )\n","\n"," # === 6. RÉSULTATS FINAUX ===\n"," best_val_acc = max(history.history['val_sparse_categorical_accuracy'])\n"," results = {\n"," 'run_id': run_id,\n"," 'config': config,\n"," 'training_time': training_time,\n"," 'final_epoch': len(history.history['loss']),\n"," 'best_val_accuracy': best_val_acc,\n"," 'test_metrics': basic_test_metrics,\n"," 'detailed_metrics': detailed_metrics,\n"," 'confusion_matrix': confusion_matrix_data,\n"," 'artifacts_dir': setup_components['artifacts_dir'],\n"," 'num_classes': setup_components['num_classes_from_model']\n"," }\n","\n"," print(f\"\\n✅ Entraînement terminé!\")\n"," print(f\"📊 Meilleure val_accuracy: {best_val_acc:.4f}\")\n"," print(f\"📊 Mean IoU: {detailed_metrics['mean_iou']:.4f}\")\n"," test_acc = basic_test_metrics.get('sparse_categorical_accuracy', 0)\n"," print(f\"🧪 Test accuracy: {test_acc:.4f}\")\n"," print(f\"📁 Artefacts: {setup_components['artifacts_dir']}\")\n"," print(f\"🔬 MLflow Run: {run_id}\")\n","\n"," return results\n","\n"," except Exception as e:\n"," mlflow.log_param(\"error\", str(e))\n"," print(f\"❌ Erreur pendant l'entraînement: {e}\")\n"," raise"],"metadata":{"id":"mWvf8oQC0UBC","executionInfo":{"status":"ok","timestamp":1755008364794,"user_tz":-120,"elapsed":66,"user":{"displayName":"Grégoire Mureau","userId":"04855910922109670788"}}},"execution_count":14,"outputs":[]},{"cell_type":"code","execution_count":15,"metadata":{"id":"ThxE2s76sTaj","executionInfo":{"status":"ok","timestamp":1755008364850,"user_tz":-120,"elapsed":54,"user":{"displayName":"Grégoire Mureau","userId":"04855910922109670788"}}},"outputs":[],"source":["def run_experiments_optimized(configs_to_test=None, experiment_name=\"OC Projet 8\"):\n"," \"\"\"\n"," Version optimisée avec système de reprise automatique + NETTOYAGE VRAM SIMPLIFIÉ\n"," \"\"\"\n","\n"," print(f\"\\n{'='*60}\")\n"," print(f\"🚀 DÉBUT DES EXPÉRIENCES - Mode: {'TEST' if TEST_MODE else 'PRODUCTION'}\")\n"," print(f\"🔬 Expérience MLflow: {experiment_name}\")\n"," print(f\"{'='*60}\")\n","\n"," # 1. Filtrage des configurations à lancer\n"," if configs_to_test is None:\n"," remaining_configs = filter_configs_to_run(experimental_configurations, experiment_name)\n","\n"," if not remaining_configs:\n"," print(\"\\n🎉 TOUTES LES EXPÉRIENCES SONT DÉJÀ TERMINÉES !\")\n"," print(\"Rien à faire. Utilisez compare_experiments_mlflow() pour voir les résultats.\")\n"," return []\n"," if TEST_MODE:\n"," configs_to_test = remaining_configs[:TEST_MAX_CONFIGS] if TEST_MAX_CONFIGS else remaining_configs\n"," print(f\"\\n🧪 Mode TEST: limitation à {len(configs_to_test)} configurations\")\n"," else:\n"," configs_to_test = remaining_configs\n"," else:\n"," completed_runs = get_completed_experiments(experiment_name)\n"," filtered_configs = []\n"," for config_idx in configs_to_test:\n"," if config_idx >= len(experimental_configurations):\n"," print(f\"⚠️ Index {config_idx} invalide, ignoré\")\n"," continue\n"," config = experimental_configurations[config_idx]\n"," expected_run_name = generate_stable_run_name(config, TEST_MODE)\n"," if expected_run_name in completed_runs:\n"," print(f\"⏭️ [{config_idx}] {config['name']} déjà terminé, ignoré\")\n"," else:\n"," filtered_configs.append(config_idx)\n"," configs_to_test = filtered_configs\n","\n"," if not configs_to_test:\n"," print(\"\\n✅ Toutes les configurations demandées sont déjà terminées !\")\n"," return []\n","\n"," # 2. Boucle d'entraînement avec nettoyage SIMPLIFIÉ\n"," print(f\"\\n🎯 ENTRAÎNEMENT DE {len(configs_to_test)} CONFIGURATIONS:\")\n","\n"," results = []\n"," for i, config_idx in enumerate(configs_to_test):\n"," config = experimental_configurations[config_idx]\n"," stable_run_name = generate_stable_run_name(config, TEST_MODE)\n","\n"," print(f\"\\n{'-'*50}\")\n"," print(f\"🧪 Expérience {i+1}/{len(configs_to_test)}: {config['name']}\")\n"," print(f\"🏷️ Run name: {stable_run_name}\")\n"," print(f\"{'-'*50}\")\n","\n"," try:\n"," # ENTRAÎNEMENT (ici artifacts_dir existe et fonctionne normalement)\n"," result = train_segmentation_model_optimized(config, run_name=stable_run_name)\n"," results.append(result)\n"," print(f\"✅ [{config_idx}] {config['name']} terminé avec succès\")\n","\n"," remaining = len(configs_to_test) - (i + 1)\n"," if remaining > 0:\n"," print(f\"🔄 {remaining} configuration(s) restante(s)\")\n","\n"," except Exception as e:\n"," print(f\"❌ Erreur avec [{config_idx}] {config['name']}: {e}\")\n","\n"," # NETTOYAGE UNIQUE : après chaque modèle (réussi ou échoué)\n"," if i < len(configs_to_test) - 1: # Pas de nettoyage après le dernier modèle\n"," print(\"🧹 Nettoyage VRAM entre modèles...\")\n"," deep_clean_memory()\n","\n"," # 3. Résumé final\n"," print(f\"\\n{'='*60}\")\n"," print(f\"📊 RÉSUMÉ DE CETTE SESSION\")\n"," print(f\"{'='*60}\")\n","\n"," if results:\n"," print(f\"✅ {len(results)} nouvelles expériences réussies\")\n","\n"," print(f\"\\n{'Modèle':<25} {'Val Acc':<10} {'Test Acc':<10} {'Mean IoU':<10} {'Temps':<8} {'Run ID':<10}\")\n"," print(\"-\" * 80)\n","\n"," for result in results:\n"," config_name = result['config']['name']\n"," val_acc = result['best_val_accuracy']\n"," test_acc = result['test_metrics'].get('sparse_categorical_accuracy', 0)\n"," mean_iou = result['detailed_metrics']['mean_iou']\n"," time_min = result['training_time'] / 60\n"," run_id = result['run_id'][:8]\n"," print(f\"{config_name:<25} {val_acc:<10.4f} {test_acc:<10.4f} {mean_iou:<10.4f} {time_min:<8.1f} {run_id:<10}\")\n","\n"," best_result = max(results, key=lambda x: x['detailed_metrics']['mean_iou'])\n"," print(f\"\\n🏆 Meilleur de cette session: {best_result['config']['name']}\")\n"," print(f\" Val Acc: {best_result['best_val_accuracy']:.4f}\")\n"," print(f\" Test Acc: {best_result['test_metrics'].get('sparse_categorical_accuracy', 0):.4f}\")\n"," print(f\" Mean IoU: {best_result['detailed_metrics']['mean_iou']:.4f}\")\n"," print(f\" MLflow Run: {best_result['run_id']}\")\n","\n"," total_remaining = filter_configs_to_run(experimental_configurations, experiment_name)\n"," if len(total_remaining) > len(configs_to_test):\n"," remaining_count = len(total_remaining) - len(configs_to_test)\n"," print(f\"\\n🔄 Il reste encore {remaining_count} configuration(s) à entraîner\")\n"," print(\"💡 Relancez le script pour continuer automatiquement\")\n"," else:\n"," print(f\"\\n🎉 TOUTES LES CONFIGURATIONS SONT MAINTENANT TERMINÉES !\")\n"," print(\"💡 Utilisez compare_experiments_local() pour voir tous les résultats\")\n"," else:\n"," print(\"❌ Aucune nouvelle expérience réussie dans cette session\")\n","\n"," return results"]},{"cell_type":"code","execution_count":16,"metadata":{"id":"SdG8VxB7v8tF","executionInfo":{"status":"ok","timestamp":1755008364905,"user_tz":-120,"elapsed":58,"user":{"displayName":"Grégoire Mureau","userId":"04855910922109670788"}}},"outputs":[],"source":["# ====================================================================\n","# COMPARAISON DES EXPÉRIENCES LOCALES\n","# ====================================================================\n","\n","def compare_experiments_local(artifacts_base_dir=\"artifacts\"):\n"," \"\"\"\n"," Compare les résultats de plusieurs expériences via les artefacts locaux\n"," Structure : artifacts/run_{run_id}/\n"," \"\"\"\n"," if not os.path.exists(artifacts_base_dir):\n"," print(\"❌ Aucun répertoire d'artefacts trouvé\")\n"," return\n","\n"," results = []\n","\n"," print(f\"🔍 Recherche des expériences dans {artifacts_base_dir}/...\")\n","\n"," # Parcourir tous les répertoires run_*\n"," run_dirs = [d for d in os.listdir(artifacts_base_dir) if d.startswith('run_')]\n","\n"," if not run_dirs:\n"," print(\"❌ Aucune expérience trouvée\")\n"," return\n","\n"," print(f\"📊 {len(run_dirs)} expériences trouvées\")\n","\n"," for run_dir in run_dirs:\n"," run_path = os.path.join(artifacts_base_dir, run_dir)\n"," config_path = os.path.join(run_path, \"model_config.json\")\n"," history_path = os.path.join(run_path, \"training_history.json\")\n"," detailed_path = os.path.join(run_path, \"detailed_metrics.json\")\n","\n"," # Vérifier que les fichiers existent\n"," if os.path.exists(config_path) and os.path.exists(history_path):\n"," try:\n"," # Charger la configuration\n"," with open(config_path, 'r') as f:\n"," config = json.load(f)\n","\n"," # Charger l'historique d'entraînement\n"," with open(history_path, 'r') as f:\n"," history = json.load(f)\n","\n"," # Charger les métriques détaillées (si disponibles)\n"," detailed_metrics = None\n"," if os.path.exists(detailed_path):\n"," with open(detailed_path, 'r') as f:\n"," detailed_metrics = json.load(f)\n","\n"," # Extraire les métriques finales\n"," final_metrics = {\n"," 'run_id': run_dir.replace('run_', ''),\n"," 'experiment': config.get('name', 'Unknown'),\n"," 'architecture': config.get('architecture', 'Unknown'),\n"," 'encoder': config.get('encoder', 'None'),\n"," 'encoder_freeze': config.get('encoder_freeze', False),\n"," 'final_train_acc': history['sparse_categorical_accuracy'][-1],\n"," 'final_val_acc': history['val_sparse_categorical_accuracy'][-1],\n"," 'final_train_loss': history['loss'][-1],\n"," 'final_val_loss': history['val_loss'][-1],\n"," 'best_val_acc': max(history['val_sparse_categorical_accuracy']),\n"," 'epochs_trained': len(history['loss']),\n"," 'mean_iou': detailed_metrics['mean_iou'] if detailed_metrics else None,\n"," 'run_path': run_path\n"," }\n"," results.append(final_metrics)\n","\n"," except Exception as e:\n"," print(f\"⚠️ Erreur lors du chargement de {run_dir}: {e}\")\n"," continue\n","\n"," # Afficher les résultats\n"," if results:\n"," print(f\"\\n🏆 COMPARAISON DES {len(results)} EXPÉRIENCES\")\n"," print(\"=\" * 100)\n","\n"," # Trier par Mean IoU si disponible, sinon par best_val_acc\n"," if any(r['mean_iou'] is not None for r in results):\n"," results_sorted = sorted(results, key=lambda x: x['mean_iou'] or 0, reverse=True)\n"," print(\"📊 Classement par Mean IoU (meilleur premier)\")\n"," else:\n"," results_sorted = sorted(results, key=lambda x: x['best_val_acc'], reverse=True)\n"," print(\"📊 Classement par Validation Accuracy (meilleur premier)\")\n","\n"," print()\n","\n"," for i, result in enumerate(results_sorted, 1):\n"," print(f\"🥇 #{i} - {result['experiment']}\")\n"," print(f\" 🏗️ Architecture: {result['architecture']}\")\n","\n"," if result['encoder'] != 'None':\n"," freeze_status = \"🧊 Frozen\" if result['encoder_freeze'] else \"🔥 Trainable\"\n"," print(f\" 🧠 Encoder: {result['encoder']} ({freeze_status})\")\n","\n"," print(f\" 📊 Meilleure Val Acc: {result['best_val_acc']:.4f}\")\n"," print(f\" 📊 Val Acc finale: {result['final_val_acc']:.4f}\")\n","\n"," if result['mean_iou'] is not None:\n"," print(f\" 🎯 Mean IoU: {result['mean_iou']:.4f}\")\n","\n"," print(f\" ⏱️ Époques: {result['epochs_trained']}\")\n"," print(f\" 🆔 Run ID: {result['run_id'][:8]}...\")\n"," print(f\" 📁 Chemin: {result['run_path']}\")\n"," print()\n","\n"," # Statistiques globales\n"," print(\"📈 STATISTIQUES GLOBALES:\")\n"," print(\"-\" * 50)\n"," avg_val_acc = sum(r['best_val_acc'] for r in results) / len(results)\n"," print(f\"🎯 Val Acc moyenne: {avg_val_acc:.4f}\")\n","\n"," if any(r['mean_iou'] is not None for r in results):\n"," valid_ious = [r['mean_iou'] for r in results if r['mean_iou'] is not None]\n"," avg_iou = sum(valid_ious) / len(valid_ious)\n"," print(f\"🎯 Mean IoU moyen: {avg_iou:.4f}\")\n","\n"," avg_epochs = sum(r['epochs_trained'] for r in results) / len(results)\n"," print(f\"⏱️ Époques moyennes: {avg_epochs:.1f}\")\n","\n"," # Architecture summary\n"," architectures = {}\n"," for r in results:\n"," arch = r['architecture']\n"," if arch not in architectures:\n"," architectures[arch] = []\n"," architectures[arch].append(r['best_val_acc'])\n","\n"," print(f\"\\n🏗️ PERFORMANCE PAR ARCHITECTURE:\")\n"," print(\"-\" * 50)\n"," for arch, accs in architectures.items():\n"," avg_acc = sum(accs) / len(accs)\n"," print(f\"{arch}: {avg_acc:.4f} (moyenne sur {len(accs)} runs)\")\n","\n"," else:\n"," print(\"❌ Aucune expérience valide trouvée\")\n","\n"," return results"]},{"cell_type":"code","execution_count":17,"metadata":{"id":"DWf1lWhyv_OI","executionInfo":{"status":"ok","timestamp":1755008364974,"user_tz":-120,"elapsed":47,"user":{"displayName":"Grégoire Mureau","userId":"04855910922109670788"}}},"outputs":[],"source":["# ====================================================================\n","# COMPARAISON DES EXPÉRIENCES LOCALES\n","# ====================================================================\n","\n","def show_best_experiment_details(artifacts_base_dir=\"artifacts\"):\n"," \"\"\"\n"," Affiche les détails de la meilleure expérience\n"," \"\"\"\n"," results = compare_experiments_local(artifacts_base_dir)\n","\n"," if not results:\n"," return\n","\n"," # Meilleure expérience (Mean IoU si disponible, sinon Val Acc)\n"," if any(r['mean_iou'] is not None for r in results):\n"," best = max(results, key=lambda x: x['mean_iou'] or 0)\n"," metric_name = \"Mean IoU\"\n"," metric_value = best['mean_iou']\n"," else:\n"," best = max(results, key=lambda x: x['best_val_acc'])\n"," metric_name = \"Validation Accuracy\"\n"," metric_value = best['best_val_acc']\n","\n"," print(f\"\\n🏆 DÉTAILS DE LA MEILLEURE EXPÉRIENCE ({metric_name})\")\n"," print(\"=\" * 60)\n"," print(f\"🎯 Nom: {best['experiment']}\")\n"," print(f\"🏗️ Architecture: {best['architecture']}\")\n","\n"," if best['encoder'] != 'None':\n"," freeze_status = \"🧊 Frozen\" if best['encoder_freeze'] else \"🔥 Trainable\"\n"," print(f\"🧠 Encoder: {best['encoder']} ({freeze_status})\")\n","\n"," print(f\"📊 {metric_name}: {metric_value:.4f}\")\n"," print(f\"📊 Validation Accuracy: {best['best_val_acc']:.4f}\")\n"," print(f\"⏱️ Époques: {best['epochs_trained']}\")\n"," print(f\"🆔 Run ID: {best['run_id']}\")\n"," print(f\"📁 Chemin des artefacts: {best['run_path']}\")\n","\n"," # Vérifier la présence du modèle\n"," model_path = os.path.join(best['run_path'], 'best_model.keras')\n"," if os.path.exists(model_path):\n"," print(f\"✅ Modèle disponible: {model_path}\")\n"," else:\n"," print(f\"❌ Modèle non trouvé: {model_path}\")"]},{"cell_type":"code","execution_count":34,"metadata":{"colab":{"base_uri":"https://localhost:8080/"},"id":"tlpWs7XqdCF5","outputId":"4e732021-35cf-43ce-cf9d-48b85d50df42","executionInfo":{"status":"ok","timestamp":1755005118660,"user_tz":-120,"elapsed":4550901,"user":{"displayName":"Grégoire Mureau","userId":"04855910922109670788"}}},"outputs":[{"output_type":"stream","name":"stdout","text":["🎯 SCRIPT D'ENTRAÎNEMENT OPTIMISÉ POUR SEGMENTATION\n","Mode: 🚀 PRODUCTION\n","🔄 SYSTÈME DE REPRISE AUTOMATIQUE ACTIVÉ\n","MLflow version: 3.2.0\n","Cityscapes dir: /content/data/cityscapes_unzipped\n","Configurations disponibles: 17\n","✅ Toutes les vérifications passées\n","\n","============================================================\n","🚀 DÉBUT DES EXPÉRIENCES - Mode: PRODUCTION\n","🔬 Expérience MLflow: OC Projet 8\n","============================================================\n","🔍 Vérification des expériences déjà terminées dans 'OC Projet 8'...\n","✅ 33 expériences terminées trouvées:\n"," - FPN_efficientnetb0_finetune_augTrue_prod\n"," - FPN_efficientnetb0_finetune_augTrue_test\n"," - FPN_efficientnetb0_frozen_augFalse_prod\n"," - FPN_efficientnetb0_frozen_augFalse_test\n"," - FPN_efficientnetb0_frozen_augTrue_prod\n"," - FPN_efficientnetb0_frozen_augTrue_test\n"," - FPN_mobilenetv2_frozen_augFalse_prod\n"," - FPN_mobilenetv2_frozen_augFalse_test\n"," - FPN_mobilenetv2_frozen_augTrue_prod\n"," - FPN_mobilenetv2_frozen_augTrue_test\n"," - FPN_resnet34_finetune_augTrue_prod\n"," - FPN_resnet34_finetune_augTrue_test\n"," - FPN_resnet34_frozen_augFalse_prod\n"," - FPN_resnet34_frozen_augFalse_test\n"," - FPN_resnet34_frozen_augTrue_prod\n"," - FPN_resnet34_frozen_augTrue_test\n"," - FPN_vanilla_augFalse_prod\n"," - FPN_vanilla_augFalse_test\n"," - FPN_vanilla_augTrue_prod\n"," - FPN_vanilla_augTrue_test\n"," - U-Net_efficientnetb0_frozen_augFalse_prod\n"," - U-Net_efficientnetb0_frozen_augFalse_test\n"," - U-Net_efficientnetb0_frozen_augTrue_prod\n"," - U-Net_efficientnetb0_frozen_augTrue_test\n"," - U-Net_mobilenetv2_finetune_augTrue_test\n"," - U-Net_mobilenetv2_frozen_augFalse_prod\n"," - U-Net_mobilenetv2_frozen_augFalse_test\n"," - U-Net_mobilenetv2_frozen_augTrue_prod\n"," - U-Net_mobilenetv2_frozen_augTrue_test\n"," - U-Net_vanilla_augFalse_prod\n"," - U-Net_vanilla_augFalse_test\n"," - U-Net_vanilla_augTrue_prod\n"," - U-Net_vanilla_augTrue_test\n","\n","📊 FILTRAGE DES CONFIGURATIONS:\n","✅ À entraîner: 1 configurations\n","⏭️ Déjà terminées: 16 configurations\n","\n","⏭️ CONFIGURATIONS IGNORÉES (déjà terminées):\n"," [0] U-Net_vanilla_augFalse → U-Net_vanilla_augFalse_prod\n"," [1] U-Net_vanilla_augTrue → U-Net_vanilla_augTrue_prod\n"," [2] FPN_vanilla_augFalse → FPN_vanilla_augFalse_prod\n"," [3] FPN_vanilla_augTrue → FPN_vanilla_augTrue_prod\n"," [4] U-Net_mobilenetv2_frozen_augFalse → U-Net_mobilenetv2_frozen_augFalse_prod\n"," [5] U-Net_mobilenetv2_frozen_augTrue → U-Net_mobilenetv2_frozen_augTrue_prod\n"," [6] U-Net_efficientnetb0_frozen_augFalse → U-Net_efficientnetb0_frozen_augFalse_prod\n"," [7] U-Net_efficientnetb0_frozen_augTrue → U-Net_efficientnetb0_frozen_augTrue_prod\n"," [8] FPN_mobilenetv2_frozen_augFalse → FPN_mobilenetv2_frozen_augFalse_prod\n"," [9] FPN_mobilenetv2_frozen_augTrue → FPN_mobilenetv2_frozen_augTrue_prod\n"," [10] FPN_efficientnetb0_frozen_augFalse → FPN_efficientnetb0_frozen_augFalse_prod\n"," [11] FPN_efficientnetb0_frozen_augTrue → FPN_efficientnetb0_frozen_augTrue_prod\n"," [12] FPN_resnet34_frozen_augFalse → FPN_resnet34_frozen_augFalse_prod\n"," [13] FPN_resnet34_frozen_augTrue → FPN_resnet34_frozen_augTrue_prod\n"," [15] FPN_efficientnetb0_finetune_augTrue → FPN_efficientnetb0_finetune_augTrue_prod\n"," [16] FPN_resnet34_finetune_augTrue → FPN_resnet34_finetune_augTrue_prod\n","\n","🚀 CONFIGURATIONS À ENTRAÎNER:\n"," [14] U-Net_mobilenetv2_finetune_augTrue → U-Net_mobilenetv2_finetune_augTrue_prod\n","\n","🎯 ENTRAÎNEMENT DE 1 CONFIGURATIONS:\n","\n","--------------------------------------------------\n","🧪 Expérience 1/1: U-Net_mobilenetv2_finetune_augTrue\n","🏷️ Run name: U-Net_mobilenetv2_finetune_augTrue_prod\n","--------------------------------------------------\n","🔬 MLflow Run ID: 61eff0ffd3554eb383ce90b7b44b31e6\n","🏷️ MLflow Run Name: U-Net_mobilenetv2_finetune_augTrue_prod\n","🚀 Mode production: batch_size=8, img_size=(224, 224)\n","\n","📊 Préparation des datasets...\n","🔍 Lecture du split Cityscapes 'train'...\n","📊 Split Cityscapes 'train': 2975 images trouvées\n","✂️ Découpage en Train/Validation (80%/20%)\n"," └─ 📚 Train final: 2380 images\n"," └─ 📊 Validation final: 595 images\n","ℹ️ Modèle vanilla - pas de preprocessing\n","🔍 Lecture du split Cityscapes 'val'...\n","📊 Split Cityscapes 'val': 500 images trouvées\n","📋 Utilisation complète comme dataset de Test\n","ℹ️ Modèle vanilla - pas de preprocessing\n","\n","🏗️ Création du modèle U-Net_mobilenetv2_finetune_augTrue...\n","📊 Nombre de classes: 8\n","📊 Paramètres: 8,048,456 total | 8,012,360 entraînables\n","\n","⚙️ Compilation du modèle avec métriques officielles...\n","🎯 MÉTRIQUES OFFICIELLES KERAS POUR SEGMENTATION SÉMANTIQUE:\n"," 🔥 LOSS: sparse_categorical_crossentropy\n"," └─ Optimise la classification pixel-par-pixel\n"," 📊 MÉTRIQUES (100% officielles Keras):\n"," ├─ SparseCategoricalAccuracy: % pixels correctement classifiés\n"," └─ MeanIoU: Intersection over Union (métrique gold standard)\n"," 🎛️ CALLBACKS:\n"," └─ Monitor: val_mean_iou (IoU = métrique de référence)\n","✅ Configuration 100% standard et fiable !\n","\n","📋 Configuration des callbacks...\n","🎛️ CALLBACKS BASÉS SUR MEAN_IOU OFFICIEL:\n"," ├─ EarlyStopping: Arrêt si val_mean_iou stagne 5 époques\n"," ├─ ReduceLROnPlateau: Réduction LR si val_mean_iou stagne 3 époques\n"," └─ ModelCheckpoint: Sauvegarde du meilleur modèle selon val_mean_iou\n","\n","🚀 Début de l'entraînement (20 epochs max)...\n","Epoch 1/20\n","\u001b[1m298/298\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 415ms/step - loss: 0.7532 - mean_iou: 0.4418 - sparse_categorical_accuracy: 0.7717"]},{"output_type":"stream","name":"stderr","text":["WARNING:absl:The `save_format` argument is deprecated in Keras 3. We recommend removing this argument as it can be inferred from the file path. Received: save_format=keras\n"]},{"output_type":"stream","name":"stdout","text":["Epoch 1: val_mean_iou improved from 0.3068773150444031 to 0.30688, saving model to artifacts/run_61eff0ffd3554eb383ce90b7b44b31e6/best_model.keras\n","\u001b[1m298/298\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m345s\u001b[0m 741ms/step - loss: 0.7525 - mean_iou: 0.4421 - sparse_categorical_accuracy: 0.7719 - val_loss: 1.3158 - val_mean_iou: 0.3069 - val_sparse_categorical_accuracy: 0.6599 - learning_rate: 0.0010\n","Epoch 2/20\n","\u001b[1m298/298\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 311ms/step - loss: 0.3834 - mean_iou: 0.6163 - sparse_categorical_accuracy: 0.8815"]},{"output_type":"stream","name":"stderr","text":["WARNING:absl:The `save_format` argument is deprecated in Keras 3. We recommend removing this argument as it can be inferred from the file path. Received: save_format=keras\n"]},{"output_type":"stream","name":"stdout","text":["Epoch 2: val_mean_iou improved from 0.3520147502422333 to 0.35201, saving model to artifacts/run_61eff0ffd3554eb383ce90b7b44b31e6/best_model.keras\n","\u001b[1m298/298\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m196s\u001b[0m 456ms/step - loss: 0.3834 - mean_iou: 0.6164 - sparse_categorical_accuracy: 0.8815 - val_loss: 1.4505 - val_mean_iou: 0.3520 - val_sparse_categorical_accuracy: 0.6578 - learning_rate: 0.0010\n","Epoch 3/20\n","\u001b[1m298/298\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 313ms/step - loss: 0.3443 - mean_iou: 0.6511 - sparse_categorical_accuracy: 0.8912"]},{"output_type":"stream","name":"stderr","text":["WARNING:absl:The `save_format` argument is deprecated in Keras 3. We recommend removing this argument as it can be inferred from the file path. Received: save_format=keras\n"]},{"output_type":"stream","name":"stdout","text":["Epoch 3: val_mean_iou improved from 0.4108622074127197 to 0.41086, saving model to artifacts/run_61eff0ffd3554eb383ce90b7b44b31e6/best_model.keras\n","\u001b[1m298/298\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m210s\u001b[0m 480ms/step - loss: 0.3442 - mean_iou: 0.6511 - sparse_categorical_accuracy: 0.8912 - val_loss: 1.2181 - val_mean_iou: 0.4109 - val_sparse_categorical_accuracy: 0.7011 - learning_rate: 0.0010\n","Epoch 4/20\n","\u001b[1m298/298\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 312ms/step - loss: 0.3168 - mean_iou: 0.6790 - sparse_categorical_accuracy: 0.9000Epoch 4: val_mean_iou did not improve from 0.41086\n","\u001b[1m298/298\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m201s\u001b[0m 460ms/step - loss: 0.3168 - mean_iou: 0.6790 - sparse_categorical_accuracy: 0.9000 - val_loss: 2.0475 - val_mean_iou: 0.2570 - val_sparse_categorical_accuracy: 0.5321 - learning_rate: 0.0010\n","Epoch 5/20\n","\u001b[1m298/298\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 314ms/step - loss: 0.3028 - mean_iou: 0.6901 - sparse_categorical_accuracy: 0.9027Epoch 5: val_mean_iou did not improve from 0.41086\n","\u001b[1m298/298\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m210s\u001b[0m 497ms/step - loss: 0.3027 - mean_iou: 0.6901 - sparse_categorical_accuracy: 0.9027 - val_loss: 1.0201 - val_mean_iou: 0.3761 - val_sparse_categorical_accuracy: 0.7217 - learning_rate: 0.0010\n","Epoch 6/20\n","\u001b[1m298/298\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 317ms/step - loss: 0.2903 - mean_iou: 0.6991 - sparse_categorical_accuracy: 0.9064"]},{"output_type":"stream","name":"stderr","text":["WARNING:absl:The `save_format` argument is deprecated in Keras 3. We recommend removing this argument as it can be inferred from the file path. Received: save_format=keras\n"]},{"output_type":"stream","name":"stdout","text":["Epoch 6: val_mean_iou improved from 0.5275560617446899 to 0.52756, saving model to artifacts/run_61eff0ffd3554eb383ce90b7b44b31e6/best_model.keras\n","\u001b[1m298/298\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m259s\u001b[0m 486ms/step - loss: 0.2903 - mean_iou: 0.6991 - sparse_categorical_accuracy: 0.9064 - val_loss: 0.6736 - val_mean_iou: 0.5276 - val_sparse_categorical_accuracy: 0.8153 - learning_rate: 0.0010\n","Epoch 7/20\n","\u001b[1m298/298\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 313ms/step - loss: 0.2720 - mean_iou: 0.7094 - sparse_categorical_accuracy: 0.9109"]},{"output_type":"stream","name":"stderr","text":["WARNING:absl:The `save_format` argument is deprecated in Keras 3. We recommend removing this argument as it can be inferred from the file path. Received: save_format=keras\n"]},{"output_type":"stream","name":"stdout","text":["Epoch 7: val_mean_iou improved from 0.5842477083206177 to 0.58425, saving model to artifacts/run_61eff0ffd3554eb383ce90b7b44b31e6/best_model.keras\n","\u001b[1m298/298\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m195s\u001b[0m 453ms/step - loss: 0.2721 - mean_iou: 0.7094 - sparse_categorical_accuracy: 0.9109 - val_loss: 0.6266 - val_mean_iou: 0.5842 - val_sparse_categorical_accuracy: 0.8313 - learning_rate: 0.0010\n","Epoch 8/20\n","\u001b[1m298/298\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 314ms/step - loss: 0.2594 - mean_iou: 0.7179 - sparse_categorical_accuracy: 0.9142Epoch 8: val_mean_iou did not improve from 0.58425\n","\u001b[1m298/298\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m201s\u001b[0m 487ms/step - loss: 0.2594 - mean_iou: 0.7179 - sparse_categorical_accuracy: 0.9141 - val_loss: 0.8479 - val_mean_iou: 0.5025 - val_sparse_categorical_accuracy: 0.7713 - learning_rate: 0.0010\n","Epoch 9/20\n","\u001b[1m298/298\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 313ms/step - loss: 0.2674 - mean_iou: 0.7141 - sparse_categorical_accuracy: 0.9124Epoch 9: val_mean_iou did not improve from 0.58425\n","\u001b[1m298/298\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m198s\u001b[0m 479ms/step - loss: 0.2674 - mean_iou: 0.7141 - sparse_categorical_accuracy: 0.9124 - val_loss: 0.8253 - val_mean_iou: 0.5051 - val_sparse_categorical_accuracy: 0.7674 - learning_rate: 0.0010\n","Epoch 10/20\n","\u001b[1m298/298\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 313ms/step - loss: 0.2552 - mean_iou: 0.7236 - sparse_categorical_accuracy: 0.9149\n","Epoch 10: ReduceLROnPlateau reducing learning rate to 0.00020000000949949026.\n","Epoch 10: val_mean_iou did not improve from 0.58425\n","\u001b[1m298/298\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m200s\u001b[0m 485ms/step - loss: 0.2552 - mean_iou: 0.7236 - sparse_categorical_accuracy: 0.9149 - val_loss: 0.7034 - val_mean_iou: 0.5600 - val_sparse_categorical_accuracy: 0.7979 - learning_rate: 0.0010\n","Epoch 11/20\n","\u001b[1m298/298\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 315ms/step - loss: 0.2262 - mean_iou: 0.7418 - sparse_categorical_accuracy: 0.9244"]},{"output_type":"stream","name":"stderr","text":["WARNING:absl:The `save_format` argument is deprecated in Keras 3. We recommend removing this argument as it can be inferred from the file path. Received: save_format=keras\n"]},{"output_type":"stream","name":"stdout","text":["Epoch 11: val_mean_iou improved from 0.660067081451416 to 0.66007, saving model to artifacts/run_61eff0ffd3554eb383ce90b7b44b31e6/best_model.keras\n","\u001b[1m298/298\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m201s\u001b[0m 488ms/step - loss: 0.2262 - mean_iou: 0.7418 - sparse_categorical_accuracy: 0.9244 - val_loss: 0.3974 - val_mean_iou: 0.6601 - val_sparse_categorical_accuracy: 0.8771 - learning_rate: 2.0000e-04\n","Epoch 12/20\n","\u001b[1m298/298\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 314ms/step - loss: 0.2136 - mean_iou: 0.7521 - sparse_categorical_accuracy: 0.9286"]},{"output_type":"stream","name":"stderr","text":["WARNING:absl:The `save_format` argument is deprecated in Keras 3. We recommend removing this argument as it can be inferred from the file path. Received: save_format=keras\n"]},{"output_type":"stream","name":"stdout","text":["Epoch 12: val_mean_iou improved from 0.6940031051635742 to 0.69400, saving model to artifacts/run_61eff0ffd3554eb383ce90b7b44b31e6/best_model.keras\n","\u001b[1m298/298\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m203s\u001b[0m 487ms/step - loss: 0.2136 - mean_iou: 0.7521 - sparse_categorical_accuracy: 0.9286 - val_loss: 0.3312 - val_mean_iou: 0.6940 - val_sparse_categorical_accuracy: 0.8965 - learning_rate: 2.0000e-04\n","Epoch 13/20\n","\u001b[1m298/298\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 311ms/step - loss: 0.2078 - mean_iou: 0.7564 - sparse_categorical_accuracy: 0.9303"]},{"output_type":"stream","name":"stderr","text":["WARNING:absl:The `save_format` argument is deprecated in Keras 3. We recommend removing this argument as it can be inferred from the file path. Received: save_format=keras\n"]},{"output_type":"stream","name":"stdout","text":["Epoch 13: val_mean_iou improved from 0.7046074867248535 to 0.70461, saving model to artifacts/run_61eff0ffd3554eb383ce90b7b44b31e6/best_model.keras\n","\u001b[1m298/298\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m258s\u001b[0m 479ms/step - loss: 0.2078 - mean_iou: 0.7564 - sparse_categorical_accuracy: 0.9303 - val_loss: 0.3175 - val_mean_iou: 0.7046 - val_sparse_categorical_accuracy: 0.9035 - learning_rate: 2.0000e-04\n","Epoch 14/20\n","\u001b[1m298/298\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 314ms/step - loss: 0.2013 - mean_iou: 0.7617 - sparse_categorical_accuracy: 0.9322"]},{"output_type":"stream","name":"stderr","text":["WARNING:absl:The `save_format` argument is deprecated in Keras 3. We recommend removing this argument as it can be inferred from the file path. Received: save_format=keras\n"]},{"output_type":"stream","name":"stdout","text":["Epoch 14: val_mean_iou improved from 0.7161905169487 to 0.71619, saving model to artifacts/run_61eff0ffd3554eb383ce90b7b44b31e6/best_model.keras\n","\u001b[1m298/298\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m200s\u001b[0m 461ms/step - loss: 0.2013 - mean_iou: 0.7617 - sparse_categorical_accuracy: 0.9322 - val_loss: 0.2965 - val_mean_iou: 0.7162 - val_sparse_categorical_accuracy: 0.9064 - learning_rate: 2.0000e-04\n","Epoch 15/20\n","\u001b[1m298/298\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 314ms/step - loss: 0.1974 - mean_iou: 0.7639 - sparse_categorical_accuracy: 0.9336"]},{"output_type":"stream","name":"stderr","text":["WARNING:absl:The `save_format` argument is deprecated in Keras 3. We recommend removing this argument as it can be inferred from the file path. Received: save_format=keras\n"]},{"output_type":"stream","name":"stdout","text":["Epoch 15: val_mean_iou improved from 0.7165768146514893 to 0.71658, saving model to artifacts/run_61eff0ffd3554eb383ce90b7b44b31e6/best_model.keras\n","\u001b[1m298/298\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m201s\u001b[0m 481ms/step - loss: 0.1975 - mean_iou: 0.7639 - sparse_categorical_accuracy: 0.9336 - val_loss: 0.2952 - val_mean_iou: 0.7166 - val_sparse_categorical_accuracy: 0.9102 - learning_rate: 2.0000e-04\n","Epoch 16/20\n","\u001b[1m298/298\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 317ms/step - loss: 0.1958 - mean_iou: 0.7665 - sparse_categorical_accuracy: 0.9338"]},{"output_type":"stream","name":"stderr","text":["WARNING:absl:The `save_format` argument is deprecated in Keras 3. We recommend removing this argument as it can be inferred from the file path. Received: save_format=keras\n"]},{"output_type":"stream","name":"stdout","text":["Epoch 16: val_mean_iou improved from 0.7194613814353943 to 0.71946, saving model to artifacts/run_61eff0ffd3554eb383ce90b7b44b31e6/best_model.keras\n","\u001b[1m298/298\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m205s\u001b[0m 487ms/step - loss: 0.1958 - mean_iou: 0.7665 - sparse_categorical_accuracy: 0.9338 - val_loss: 0.2903 - val_mean_iou: 0.7195 - val_sparse_categorical_accuracy: 0.9085 - learning_rate: 2.0000e-04\n","Epoch 17/20\n","\u001b[1m298/298\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 316ms/step - loss: 0.1893 - mean_iou: 0.7695 - sparse_categorical_accuracy: 0.9359"]},{"output_type":"stream","name":"stderr","text":["WARNING:absl:The `save_format` argument is deprecated in Keras 3. We recommend removing this argument as it can be inferred from the file path. Received: save_format=keras\n"]},{"output_type":"stream","name":"stdout","text":["Epoch 17: val_mean_iou improved from 0.7226355075836182 to 0.72264, saving model to artifacts/run_61eff0ffd3554eb383ce90b7b44b31e6/best_model.keras\n","\u001b[1m298/298\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m196s\u001b[0m 459ms/step - loss: 0.1894 - mean_iou: 0.7695 - sparse_categorical_accuracy: 0.9359 - val_loss: 0.2893 - val_mean_iou: 0.7226 - val_sparse_categorical_accuracy: 0.9092 - learning_rate: 2.0000e-04\n","Epoch 18/20\n","\u001b[1m298/298\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 319ms/step - loss: 0.1881 - mean_iou: 0.7731 - sparse_categorical_accuracy: 0.9363"]},{"output_type":"stream","name":"stderr","text":["WARNING:absl:The `save_format` argument is deprecated in Keras 3. We recommend removing this argument as it can be inferred from the file path. Received: save_format=keras\n"]},{"output_type":"stream","name":"stdout","text":["Epoch 18: val_mean_iou improved from 0.7297743558883667 to 0.72977, saving model to artifacts/run_61eff0ffd3554eb383ce90b7b44b31e6/best_model.keras\n","\u001b[1m298/298\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m204s\u001b[0m 491ms/step - loss: 0.1881 - mean_iou: 0.7731 - sparse_categorical_accuracy: 0.9363 - val_loss: 0.2871 - val_mean_iou: 0.7298 - val_sparse_categorical_accuracy: 0.9136 - learning_rate: 2.0000e-04\n","Epoch 19/20\n","\u001b[1m298/298\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 315ms/step - loss: 0.1838 - mean_iou: 0.7756 - sparse_categorical_accuracy: 0.9375Epoch 19: val_mean_iou did not improve from 0.72977\n","\u001b[1m298/298\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m259s\u001b[0m 484ms/step - loss: 0.1839 - mean_iou: 0.7756 - sparse_categorical_accuracy: 0.9375 - val_loss: 0.3005 - val_mean_iou: 0.7193 - val_sparse_categorical_accuracy: 0.9070 - learning_rate: 2.0000e-04\n","Epoch 20/20\n","\u001b[1m298/298\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 315ms/step - loss: 0.1812 - mean_iou: 0.7760 - sparse_categorical_accuracy: 0.9385"]},{"output_type":"stream","name":"stderr","text":["WARNING:absl:The `save_format` argument is deprecated in Keras 3. We recommend removing this argument as it can be inferred from the file path. Received: save_format=keras\n"]},{"output_type":"stream","name":"stdout","text":["Epoch 20: val_mean_iou improved from 0.7328457832336426 to 0.73285, saving model to artifacts/run_61eff0ffd3554eb383ce90b7b44b31e6/best_model.keras\n","\u001b[1m298/298\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m202s\u001b[0m 485ms/step - loss: 0.1813 - mean_iou: 0.7760 - sparse_categorical_accuracy: 0.9385 - val_loss: 0.2888 - val_mean_iou: 0.7328 - val_sparse_categorical_accuracy: 0.9124 - learning_rate: 2.0000e-04\n","Restoring model weights from the end of the best epoch: 20.\n","⏱️ Temps d'entraînement: 4346.37s\n","\n","📊 Évaluation avec métriques officielles Keras...\n","\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m41s\u001b[0m 485ms/step - loss: 0.2914 - mean_iou: 0.7216 - sparse_categorical_accuracy: 0.9103\n","✅ Test loss (officiel): 0.3201\n","✅ Test sparse_categorical_accuracy (officiel): 0.9012\n","✅ Test mean_iou (officiel): 0.7130\n","\n","🔍 Calcul des métriques détaillées par classe...\n"," 📊 Analyse détaillée par classe...\n"," Batch 1 traité...\n"," Batch 11 traité...\n"," Batch 21 traité...\n"," Batch 31 traité...\n"," Batch 41 traité...\n"," Batch 51 traité...\n"," Batch 61 traité...\n"," Total d'images évaluées: 500\n","\n","📊 ANALYSE DÉTAILLÉE PAR CLASSE:\n","----------------------------------------------------------------------\n","Classe IoU Vrais % Prédits % \n","----------------------------------------------------------------------\n","flat 0.9257 38.01 39.03 \n","human 0.4706 1.32 1.74 \n","vehicle 0.7988 7.08 7.28 \n","construction 0.7930 20.49 21.42 \n","object 0.3035 2.05 1.71 \n","nature 0.8403 15.84 15.37 \n","sky 0.8823 2.94 3.08 \n","void 0.6897 12.28 10.38 \n","----------------------------------------------------------------------\n","MEAN IoU (calc) 0.7130 \n","💡 Le Mean IoU officiel (métrique Keras) fait référence\n","\n","🔥 Génération de la matrice de confusion...\n","📊 Calcul de la matrice de confusion...\n","✅ Matrice de confusion sauvegardée : artifacts/run_61eff0ffd3554eb383ce90b7b44b31e6/confusion_matrix.png\n","\n","📈 Création des visualisations...\n","\n","💾 Sauvegarde des artefacts...\n","\n","✅ Entraînement terminé!\n","📊 Meilleure val_accuracy: 0.9136\n","📊 Mean IoU: 0.7130\n","🧪 Test accuracy: 0.9012\n","📁 Artefacts: artifacts/run_61eff0ffd3554eb383ce90b7b44b31e6\n","🔬 MLflow Run: 61eff0ffd3554eb383ce90b7b44b31e6\n","🏃 View run U-Net_mobilenetv2_finetune_augTrue_prod at: https://mlflow.greg-madman-nas.duckdns.org/#/experiments/3/runs/61eff0ffd3554eb383ce90b7b44b31e6\n","🧪 View experiment at: https://mlflow.greg-madman-nas.duckdns.org/#/experiments/3\n","✅ [14] U-Net_mobilenetv2_finetune_augTrue terminé avec succès\n","\n","============================================================\n","📊 RÉSUMÉ DE CETTE SESSION\n","============================================================\n","✅ 1 nouvelles expériences réussies\n","\n","Modèle Val Acc Test Acc Mean IoU Temps Run ID \n","--------------------------------------------------------------------------------\n","U-Net_mobilenetv2_finetune_augTrue 0.9136 0.9012 0.7130 72.4 61eff0ff \n","\n","🏆 Meilleur de cette session: U-Net_mobilenetv2_finetune_augTrue\n"," Val Acc: 0.9136\n"," Test Acc: 0.9012\n"," Mean IoU: 0.7130\n"," MLflow Run: 61eff0ffd3554eb383ce90b7b44b31e6\n","🔍 Vérification des expériences déjà terminées dans 'OC Projet 8'...\n","✅ 34 expériences terminées trouvées:\n"," - FPN_efficientnetb0_finetune_augTrue_prod\n"," - FPN_efficientnetb0_finetune_augTrue_test\n"," - FPN_efficientnetb0_frozen_augFalse_prod\n"," - FPN_efficientnetb0_frozen_augFalse_test\n"," - FPN_efficientnetb0_frozen_augTrue_prod\n"," - FPN_efficientnetb0_frozen_augTrue_test\n"," - FPN_mobilenetv2_frozen_augFalse_prod\n"," - FPN_mobilenetv2_frozen_augFalse_test\n"," - FPN_mobilenetv2_frozen_augTrue_prod\n"," - FPN_mobilenetv2_frozen_augTrue_test\n"," - FPN_resnet34_finetune_augTrue_prod\n"," - FPN_resnet34_finetune_augTrue_test\n"," - FPN_resnet34_frozen_augFalse_prod\n"," - FPN_resnet34_frozen_augFalse_test\n"," - FPN_resnet34_frozen_augTrue_prod\n"," - FPN_resnet34_frozen_augTrue_test\n"," - FPN_vanilla_augFalse_prod\n"," - FPN_vanilla_augFalse_test\n"," - FPN_vanilla_augTrue_prod\n"," - FPN_vanilla_augTrue_test\n"," - U-Net_efficientnetb0_frozen_augFalse_prod\n"," - U-Net_efficientnetb0_frozen_augFalse_test\n"," - U-Net_efficientnetb0_frozen_augTrue_prod\n"," - U-Net_efficientnetb0_frozen_augTrue_test\n"," - U-Net_mobilenetv2_finetune_augTrue_prod\n"," - U-Net_mobilenetv2_finetune_augTrue_test\n"," - U-Net_mobilenetv2_frozen_augFalse_prod\n"," - U-Net_mobilenetv2_frozen_augFalse_test\n"," - U-Net_mobilenetv2_frozen_augTrue_prod\n"," - U-Net_mobilenetv2_frozen_augTrue_test\n"," - U-Net_vanilla_augFalse_prod\n"," - U-Net_vanilla_augFalse_test\n"," - U-Net_vanilla_augTrue_prod\n"," - U-Net_vanilla_augTrue_test\n","\n","📊 FILTRAGE DES CONFIGURATIONS:\n","✅ À entraîner: 0 configurations\n","⏭️ Déjà terminées: 17 configurations\n","\n","⏭️ CONFIGURATIONS IGNORÉES (déjà terminées):\n"," [0] U-Net_vanilla_augFalse → U-Net_vanilla_augFalse_prod\n"," [1] U-Net_vanilla_augTrue → U-Net_vanilla_augTrue_prod\n"," [2] FPN_vanilla_augFalse → FPN_vanilla_augFalse_prod\n"," [3] FPN_vanilla_augTrue → FPN_vanilla_augTrue_prod\n"," [4] U-Net_mobilenetv2_frozen_augFalse → U-Net_mobilenetv2_frozen_augFalse_prod\n"," [5] U-Net_mobilenetv2_frozen_augTrue → U-Net_mobilenetv2_frozen_augTrue_prod\n"," [6] U-Net_efficientnetb0_frozen_augFalse → U-Net_efficientnetb0_frozen_augFalse_prod\n"," [7] U-Net_efficientnetb0_frozen_augTrue → U-Net_efficientnetb0_frozen_augTrue_prod\n"," [8] FPN_mobilenetv2_frozen_augFalse → FPN_mobilenetv2_frozen_augFalse_prod\n"," [9] FPN_mobilenetv2_frozen_augTrue → FPN_mobilenetv2_frozen_augTrue_prod\n"," [10] FPN_efficientnetb0_frozen_augFalse → FPN_efficientnetb0_frozen_augFalse_prod\n"," [11] FPN_efficientnetb0_frozen_augTrue → FPN_efficientnetb0_frozen_augTrue_prod\n"," [12] FPN_resnet34_frozen_augFalse → FPN_resnet34_frozen_augFalse_prod\n"," [13] FPN_resnet34_frozen_augTrue → FPN_resnet34_frozen_augTrue_prod\n"," [14] U-Net_mobilenetv2_finetune_augTrue → U-Net_mobilenetv2_finetune_augTrue_prod\n"," [15] FPN_efficientnetb0_finetune_augTrue → FPN_efficientnetb0_finetune_augTrue_prod\n"," [16] FPN_resnet34_finetune_augTrue → FPN_resnet34_finetune_augTrue_prod\n","\n","🎉 TOUTES LES CONFIGURATIONS SONT MAINTENANT TERMINÉES !\n","💡 Utilisez compare_experiments_local() pour voir tous les résultats\n","\n","🎉 Script terminé avec succès!\n"]}],"source":["# ====================================================================\n","# POINT D'ENTRÉE PRINCIPAL\n","# ====================================================================\n","\n","if __name__ == \"__main__\":\n"," print(\"🎯 SCRIPT D'ENTRAÎNEMENT OPTIMISÉ POUR SEGMENTATION\")\n"," print(f\"Mode: {'🧪 TEST' if TEST_MODE else '🚀 PRODUCTION'}\")\n"," print(\"🔄 SYSTÈME DE REPRISE AUTOMATIQUE ACTIVÉ\")\n","\n"," # Vérifications préliminaires\n"," try:\n"," print(f\"MLflow version: {mlflow.__version__}\")\n"," print(f\"Cityscapes dir: {cityscapes_dir}\")\n"," print(f\"Configurations disponibles: {len(experimental_configurations)}\")\n","\n"," assert callable(create_model_from_config), \"create_model_from_config non disponible\"\n"," assert callable(prepare_cityscapes_dataset), \"prepare_cityscapes_dataset non disponible\"\n","\n"," print(\"✅ Toutes les vérifications passées\")\n","\n"," # Lancer les expériences avec système de reprise\n"," results = run_experiments_optimized()\n","\n"," print(\"\\n🎉 Script terminé avec succès!\")\n","\n"," except Exception as e:\n"," print(f\"❌ Erreur fatale: {e}\")\n"," print(\"Vérifications:\")\n"," print(\"1. Variables cityscapes_dir et experimental_configurations définies ?\")\n"," print(\"2. Fonctions create_model_from_config, prepare_cityscapes_dataset disponibles ?\")\n"," print(\"3. MLflow configuré ?\")\n"," print(\"4. Données Cityscapes présentes ?\")"]},{"cell_type":"markdown","metadata":{"id":"6gJK8H-YRwRN"},"source":["## Analyse des expériences"]},{"cell_type":"markdown","metadata":{"id":"erK7RXCV7HOz"},"source":["### Comparaison des experiences"]},{"cell_type":"code","execution_count":25,"metadata":{"id":"LQqdr5KUTWZW","executionInfo":{"status":"ok","timestamp":1755008394020,"user_tz":-120,"elapsed":28,"user":{"displayName":"Grégoire Mureau","userId":"04855910922109670788"}}},"outputs":[],"source":["def get_completed_configs_info(experimental_configurations, experiment_name=\"OC Projet 8\"):\n"," \"\"\"\n"," Réutilise la logique de filter_configs_to_run mais retourne les configs terminées\n"," \"\"\"\n"," completed_runs = get_completed_experiments(experiment_name)\n","\n"," if not completed_runs:\n"," return []\n","\n"," # Même logique que filter_configs_to_run mais on garde les terminées\n"," completed_configs = []\n","\n"," for idx, config in enumerate(experimental_configurations):\n"," expected_run_name = generate_stable_run_name(config, TEST_MODE)\n","\n"," if expected_run_name in completed_runs: # Condition inversée\n"," completed_configs.append((idx, config, expected_run_name))\n","\n"," return completed_configs\n","\n","def compare_experiments_mlflow(experiment_name=\"OC Projet 8\", show_detailed_results=False, show_plots=True):\n"," \"\"\"\n"," Compare les expériences terminées - version ultra-simple\n","\n"," Args:\n"," experiment_name (str): Nom de l'expérience MLflow\n"," show_detailed_results (bool): Si True, affiche et retourne la liste complète des résultats\n"," show_plots (bool): Si True, génère les diagrammes en barres des métriques\n"," \"\"\"\n"," # Récupérer les configs terminées (réutilise la logique existante)\n"," completed_configs = get_completed_configs_info(experimental_configurations, experiment_name)\n","\n"," if not completed_configs:\n"," print(\"❌ Aucune expérience terminée trouvée\")\n"," return []\n","\n"," # Récupérer le mapping run_name -> run_id depuis MLflow\n"," experiment = mlflow.get_experiment_by_name(experiment_name)\n"," runs_df = mlflow.search_runs(\n"," experiment_ids=[experiment.experiment_id],\n"," filter_string=\"status = 'FINISHED'\",\n"," output_format=\"pandas\"\n"," )\n","\n"," run_name_to_id = {\n"," run.get('tags.mlflow.runName', ''): run['run_id']\n"," for _, run in runs_df.iterrows()\n"," }\n","\n"," print(f\"📊 Analyse de {len(completed_configs)} configurations terminées\")\n","\n"," results = []\n","\n"," for config_idx, config, run_name in completed_configs:\n"," if run_name not in run_name_to_id:\n"," continue\n","\n"," run_id = run_name_to_id[run_name]\n","\n"," try:\n"," # Récupérer les métriques directement depuis MLflow\n"," run = mlflow.get_run(run_id)\n"," metrics = run.data.metrics\n","\n"," # Récupérer l'historique d'entraînement pour certaines infos\n"," import tempfile\n"," with tempfile.TemporaryDirectory() as temp_dir:\n"," try:\n"," history_path = mlflow.artifacts.download_artifacts(\n"," run_id=run_id, artifact_path=\"training_history.json\", dst_path=temp_dir\n"," )\n"," with open(history_path, 'r') as f:\n"," history = json.load(f)\n"," best_val_acc = max(history['val_sparse_categorical_accuracy'])\n"," epochs_trained = len(history['loss'])\n"," except:\n"," # Fallback si l'artefact n'est pas disponible\n"," best_val_acc = metrics.get('val_sparse_categorical_accuracy', 'N/A')\n"," epochs_trained = 'N/A'\n","\n"," # Construire le résultat avec les métriques de test depuis MLflow\n"," result = {\n"," 'config_idx': config_idx,\n"," 'experiment': config['name'],\n"," 'architecture': config.get('architecture', 'Unknown'),\n"," 'encoder': config.get('encoder', 'None'),\n"," 'best_val_acc': best_val_acc,\n"," 'test_mean_iou': metrics.get('test_mean_iou', 'N/A'),\n"," 'test_accuracy': metrics.get('test_sparse_categorical_accuracy', 'N/A'),\n"," 'epochs_trained': epochs_trained,\n"," 'run_name': run_name,\n"," 'run_id': run_id,\n"," 'full_config': config # Toujours inclure la config complète\n"," }\n","\n"," results.append(result)\n","\n"," except Exception as e:\n"," print(f\"⚠️ Erreur pour [{config_idx}] {config['name']}: {e}\")\n"," continue\n","\n"," # Affichage\n"," if results:\n"," # Trier par test_mean_iou si disponible, sinon par best_val_acc\n"," def sort_key(x):\n"," if x['test_mean_iou'] != 'N/A':\n"," return x['test_mean_iou']\n"," return x['best_val_acc']\n","\n"," results_sorted = sorted(results, key=sort_key, reverse=True)\n","\n"," print(f\"\\n🏆 TOP {len(results)} EXPÉRIENCES TERMINÉES\")\n"," print(\"=\" * 100)\n","\n"," for i, result in enumerate(results_sorted, 1):\n"," print(f\"🥇 #{i} - [{result['config_idx']}] {result['experiment']}\")\n"," print(f\" 🏗️ {result['architecture']} + {result['encoder']}\")\n","\n"," # Afficher les métriques de test\n"," test_iou = result['test_mean_iou']\n"," test_acc = result['test_accuracy']\n","\n"," if test_iou != 'N/A':\n"," print(f\" 🎯 Test Mean IoU: {test_iou:.4f}\", end=\"\")\n"," else:\n"," print(f\" 🎯 Test Mean IoU: N/A\", end=\"\")\n","\n"," if test_acc != 'N/A':\n"," print(f\" | Test Acc: {test_acc:.4f}\")\n"," else:\n"," print(f\" | Test Acc: N/A\")\n","\n"," print(f\" 🆔 {result['run_id'][:8]}...\")\n","\n"," # Afficher la configuration complète si demandé\n"," if show_detailed_results:\n"," print(f\" 📋 Configuration complète:\")\n"," import pprint\n"," pprint.pprint(result['full_config'], indent=6)\n","\n"," print()\n","\n"," # Génération des diagrammes si demandé\n"," if show_plots and results:\n"," generate_comparison_plots(results_sorted)\n","\n"," # Retourner les résultats seulement si explicitement demandé\n"," if show_detailed_results:\n"," return results\n"," else:\n"," print(f\"\\n✅ Analyse terminée. Utilisez show_detailed_results=True pour obtenir les données détaillées.\")\n"," return None\n","\n","\n","def generate_comparison_plots(results):\n"," \"\"\"\n"," Génère des diagrammes en barres pour comparer les métriques de test\n","\n"," Args:\n"," results (list): Liste des résultats triés\n"," \"\"\"\n"," import matplotlib.pyplot as plt\n"," import numpy as np\n","\n"," # Filtrer les résultats qui ont des métriques de test valides\n"," valid_results_iou = [r for r in results if r['test_mean_iou'] != 'N/A']\n"," valid_results_acc = [r for r in results if r['test_accuracy'] != 'N/A']\n","\n"," # Configuration des graphiques\n"," plt.style.use('seaborn-v0_8' if 'seaborn-v0_8' in plt.style.available else 'default')\n","\n"," # Créer une figure avec 2 sous-graphiques\n"," fig, (ax1, ax2) = plt.subplots(1, 2, figsize=(20, 8))\n","\n"," # Graphique 1: Test Mean IoU\n"," if valid_results_iou:\n"," experiment_names = [r['experiment'] for r in valid_results_iou]\n"," iou_values = [r['test_mean_iou'] for r in valid_results_iou]\n","\n"," # Créer des labels courts pour les noms d'expériences\n"," short_labels = []\n"," for name in experiment_names:\n"," # Raccourcir les noms longs\n"," if len(name) > 25:\n"," parts = name.split('_')\n"," short_name = f\"{parts[0]}_{parts[1] if len(parts) > 1 else ''}...\"\n"," short_labels.append(short_name)\n"," else:\n"," short_labels.append(name)\n","\n"," colors = plt.cm.viridis(np.linspace(0, 1, len(iou_values)))\n"," bars1 = ax1.bar(range(len(iou_values)), iou_values, color=colors, alpha=0.8)\n","\n"," ax1.set_title('Test Mean IoU par Expérience', fontsize=14, fontweight='bold')\n"," ax1.set_xlabel('Expériences', fontsize=12)\n"," ax1.set_ylabel('Mean IoU', fontsize=12)\n"," ax1.set_xticks(range(len(short_labels)))\n"," ax1.set_xticklabels(short_labels, rotation=45, ha='right')\n"," ax1.grid(True, alpha=0.3)\n","\n"," # Ajouter les valeurs sur les barres\n"," for bar, value in zip(bars1, iou_values):\n"," height = bar.get_height()\n"," ax1.text(bar.get_x() + bar.get_width()/2., height + 0.001,\n"," f'{value:.4f}', ha='center', va='bottom', fontsize=10)\n"," else:\n"," ax1.text(0.5, 0.5, 'Aucune donnée Mean IoU disponible',\n"," ha='center', va='center', transform=ax1.transAxes, fontsize=14)\n"," ax1.set_title('Test Mean IoU par Expérience', fontsize=14, fontweight='bold')\n","\n"," # Graphique 2: Test Accuracy\n"," if valid_results_acc:\n"," experiment_names = [r['experiment'] for r in valid_results_acc]\n"," acc_values = [r['test_accuracy'] for r in valid_results_acc]\n","\n"," # Créer des labels courts pour les noms d'expériences\n"," short_labels = []\n"," for name in experiment_names:\n"," if len(name) > 25:\n"," parts = name.split('_')\n"," short_name = f\"{parts[0]}_{parts[1] if len(parts) > 1 else ''}...\"\n"," short_labels.append(short_name)\n"," else:\n"," short_labels.append(name)\n","\n"," colors = plt.cm.plasma(np.linspace(0, 1, len(acc_values)))\n"," bars2 = ax2.bar(range(len(acc_values)), acc_values, color=colors, alpha=0.8)\n","\n"," ax2.set_title('Test Accuracy par Expérience', fontsize=14, fontweight='bold')\n"," ax2.set_xlabel('Expériences', fontsize=12)\n"," ax2.set_ylabel('Accuracy', fontsize=12)\n"," ax2.set_xticks(range(len(short_labels)))\n"," ax2.set_xticklabels(short_labels, rotation=45, ha='right')\n"," ax2.grid(True, alpha=0.3)\n","\n"," # Ajouter les valeurs sur les barres\n"," for bar, value in zip(bars2, acc_values):\n"," height = bar.get_height()\n"," ax2.text(bar.get_x() + bar.get_width()/2., height + 0.001,\n"," f'{value:.4f}', ha='center', va='bottom', fontsize=10)\n"," else:\n"," ax2.text(0.5, 0.5, 'Aucune donnée Accuracy disponible',\n"," ha='center', va='center', transform=ax2.transAxes, fontsize=14)\n"," ax2.set_title('Test Accuracy par Expérience', fontsize=14, fontweight='bold')\n","\n"," plt.tight_layout()\n"," plt.show()\n","\n"," print(f\"📊 Diagrammes générés pour {len(valid_results_iou)} expériences (IoU) et {len(valid_results_acc)} expériences (Accuracy)\")"]},{"cell_type":"code","execution_count":26,"metadata":{"id":"fcl3B8x7PPrc","colab":{"base_uri":"https://localhost:8080/","height":1000,"referenced_widgets":["6a2e155bbb79498a86a9651abf25989d","f1757c6193dc47a3adfc732d78478f47","8d0f263257944fea863c5bb8a7c23093","6cd5b8ad3b584f7f837475371bf4a4c5","45b8aa12303c446b999ea9c3c2ab1f2a","8c9db1a5fc8f48bbafd5c6eae6cf3f85","407276af96c045fd852e858d03653329","0bc6887908f54cdb962ffa6a136a1619","92785f12b1284ca4bfb517a89acff47b","de3b5dceb033475e917806eb2740212e","233060dd87f24a4786edd722412d7f09","c8579f95a6cb473da8381368739b36df","4b7cd64207b24e1ebf105085a7bc909c","3a8379bf097944a7823b80d15e513d70","171d955212f64caeb7ea500dc0bfdaf0","bdf266e8a3bd4402ad4403b84da3053c","cc7ccd07ecfc48fcb28b15a04a5174a7","e61820d5891b47128b53fa8fa5ed8638","bae039121a3143de98dc60efb3e9ed14","ac3f93b378174cb7991460825ed48b7f","ce8bb032b28c4896ad805430ac1c6509","2c8a0f7b1e434615afaf4a1e51307f37","790fb9cc1fa64a19b357a66e7c5f64dc","60b4ba2a30b840afa830bbe029695ad0","c6eedfc5eb3347758274b5e2cb72a435","d205a4e7e2a14896ae756939c8abbafc","f2d7f5f153134de8b1d2940e5ea91424","f4d060ed2c9443a6824179c8c2f747af","4c9577c9bb9842d2860a777678903fbc","15f392891c4c4efdb658c29f8e6d986d","107b71bd8800421eb90a61fb46ce727f","4be89f9f4cfe465291b28e4c437d72f3","a473fa7827064b3aaf27f0c79a74abdc","573fbfcf254b4231ac218222cf166bd7","e0c4426574b5444caa3f3076409a8da2","4d1739b825c745f7ad97ba42ec1f1524","1195fb90d760415e8a47824938f189d1","2a103028dcc54d798adff13f2f411497","cadbf166edca428dac8e99b8ab52d27c","8dd830023f774e28b650d54a68be863f","2f2c5f02368946569444daa3db213d9e","23dfd1c3327e4d27877fca6091c3a768","f8f343a67f124aa9ba7e41201de9f5aa","fceacc6bcb694b099767c6b9e8eab1c7","0b5dd0fc844249e18f8fc369451b9e57","f83414fe4d7644aab33e391e418f919a","3a59562d047348f4bce216be1df20985","ef5809659d1344d798df4eaaddef7fe1","97d56496bd7e41068490330318e2548e","0a5875cb88d648b88540ccad1638cef4","626f2a7e404c429097ff6f827335a08d","207836149fc1456794c336c945e21e48","2033b77d087649298d3eaefff36231bf","7aa7fa0cfaff46cf897ce46a821c7d1f","49adb052e6d74fbea07c8612ca7834e9","0f4b6a7074a64c53a8c2144dcb5d10d2","253c2715bb39452a94f720dafece2e3c","abd19746f5474d79a9673e418e780ea1","3b0020b599364d0e827778e221f0c836","b7dcd45fb19c4c2797f597bca0520169","e1905276708a4885a127c3c32735028d","71bebac0f80241c8be262b4f6b85675f","a7850fa848cf4cbba94de61c4feb12c4","a32c906aefcf4eb68d93fd381a1430cf","01604a2728ec4e419c040365d867c11c","1a5cc47606e7414398175ffa770c319d","a81a7abf31f24bcbafebe038fad72d04","f25c411df73243c98fc383e81880470b","71df0041991e4f7eab75f03c5520d370","dfb9d63c2d084cefa1b82e8e1b43c9e1","c8f2e31860314c9ba70903a5e57a8d40","3c467f55e04f419dac3a239c21533c9d","8852c726203641d49cc24056d3d36265","9610052f64ee49d690e941495fad9d8a","e351892331c546708c18ab5f0a9a9abe","6c8cff7d4f064c8e88788f09e7df33cc","08582f2e68ee428183c3041f93e46fdd","1b7255beace348178704bef7a0fe7e08","da650e3512734e2d8241c3aabdba08bd","deedd69deab74fc090c5217119b49b4b","43ac57b840de4bdc89c47880f4eb776d","357cfaf53a8a4e0fbfd79a3c091d7820","bd2d98fc8b874732b0c480103bff8c36","29f98c5eb1954caf812a84ec055596cd","482fba717d5e48c88841f7208427b441","6866047084dd4ab089862831597d34be","2ed8d6aa0dca4a0b93e189e36818fd69","60597cb773644b2484049113b4258ea0","1288fd7dba28488e8acfab0f1a2e8956","18e2b7048afc4f72911e4981e97857a2","03c268e44c6941ad8c5675b9718c2503","082c104effd64bdcace48449f7206a52","ffe3598bb370435ea910b27fc005cf3c","a4f53135ce0949adba8e00e30d24df75","85ad9cada2564958acdcc5d7bdf3d8dd","253c7c750170409d95c922e0a50137c7","c0e372c2bc274c07a5c21dd3e6565c4a","0dfcd80005ac443ba6d3f50bdb92dfe0","6860d56adb484d26b9a600cf64c837ba","702cc276b9204f318b90e19fb155d8ab","c16f87e486e642aa89da38ddf43c67d1","787001192446437992390b8bfa02fb5d","be411d56ae8a44b495d7b1689aca4f4d","b683d5e9718844a8b7489ce944c924c6","45a368a1a4b04f73b43cf2454a56f67f","49bc32445c994ca5a738d9895d52e8b8","852e8d7b5b904b63b7e0c26d19f33236","3df6fc1b79ff4e3bbbeab287987a8864","27f1a4ae0a0f4ddd94eb5c567d7519e1","497d63d827b34ea2874487fb3b459dba","c841c2d981b4418a90c7c6126e57b60e","c9318a79ec2646408342fd7961ee6608","d4d80c9a41924d05b52df669aec3e170","252599c8e98b452e8538dc2aba0eeac1","dae1feecd17f48ae9392c9e498ba4d72","44e74ea4e22f4240afb96829c7c3abc2","8dc3b46623d340bfa4653d46d10def82","09b7f0cacea54523b1a70285a73e5b74","686c1eb6f22e4dabbc9778950a1eced7","b0e7761e7191423990735d62d7422d24","53508f709b6a4682840002a0a4d5802a","7d2545733b9b43a6a6e10bcd794a1d88","72255fe7370748279f701dc49b9add76","5d1f12df56664e3993ddb609829e5317","f782da7721af4570838d2dcdef3ef1e6","d5cb1acba70b45449f898619e7894995","f29f4922d3a8488389d8286bd8058d2e","639b83e0f1774acfa32f45c1b3b8d452","2279d48c035f4f86a0a6c4d181f94f39","9ea29434d97846ea90008754b4cc88fd","d5c0ca1b4447481090322081a14a5672","c6623f08b33a4b8f99d2b19e619ced44","593c028904024b819038fb75554a3ce9","caefb20bc0ab40689e83d8a460123e60","764eb4520cf84c8e840dd8a4d87b9781","8dd28c53b2424836b0ec3618c4564c21","b30ee7473f0e4821be7e02717f57c9fd","eb3b4f1189df4fd6ad1863fdd248bace","795feeccb71449679e5e48f7bc0bfdc9","ab996259d78a41d18113a286af0807c8","0fef5b65d42c4c9eb54c607aa1f533dd","abeb6f32362e46bdacc733cc8207dd74","623e6144c6b449d6ab3481c7dcbad3b6","e07785945add4f2bae2351ef39b500fa","7a747a0947ae4d9b91d495c553e4178f","3983b2f0a1e34014a6f698c8b5de2e89","dea8fa739aaa4850b715af20b7165451","f4a170eaa61c400fa59900f9179c6fe8","37da640c239c432cb2be9c6498d4de16","4d56da05aa29409986c3a3a2265a77be","64e05402cf8e4b79a62249aa7fa91c64","3b52cd8baa944053bbd082e80d2d1ee9","41d997aac92e47c78fdbd077ef88e862","37f459cf793c4319b7e0741ecac3d900","9861231e8d404e338a42311abe585001","ecefa3ff534e4c62bcdc91d5bce99b7b","0208a1f0d69c4a1292a309122e69bd1e","ff31838e68e74a9d9f4d2fb599101f83","a9a22162cdc441beabd2210cd98a1b6e","205f5b95acb8440dab6b77cbd4ae90cf","89f0b792653946e682bc6bfe431375c5","83978bd505494886a6be7594e9a59c5b","b1d83e4c99054daf9f56c408c4cc5310","95fdc91772c748479b3e364c38e0a03a","10ce15520aec46f6bfe8b12be9eb53d3","e36b21c28c9548e8ba4acd4165b96f64","8f29364c251b496a9515f610f99a8f6a","547fce071e0640f79220428be87dd9be","8c23cb823eb14ba8a3bbf35f76f2d8f6","efc6432d6586407fa4b5a29303108fa8","88839cafd96e4c1796bf981b11503ce0","63df7199831846aab8054101a3a81da3","18328a92674d417f9568519559d2788a","91254d62949d4a69954dbfd4b4875158","58a46397428b474396fd0cac891570e3","882096450080477ab1b8262506a2a546","fd7e8031b15940aebdc1a12718b89a8f","fd2ee43fae38435eacd0ead8be3af856","43061552c25d41f4a5fc0c8428e27a62","8cc46372b8b14bc6ba78aea2a129e3d8","2e24cc9a551245318d4fdbce788e64cd","9e52452152694ac28ce431f2311024c8","ab1d062f109b475d8ef2613fbe7ad729","a29e94dd53a24f0e8cabe84229bef23d","5f03aacf9f0e42488eb262dc3856b56c","d193689b2c9f432297c88d39d90b5991","2bf0e548dd95411481fc1778ad685bca"]},"executionInfo":{"status":"ok","timestamp":1755008415666,"user_tz":-120,"elapsed":21480,"user":{"displayName":"Grégoire Mureau","userId":"04855910922109670788"}},"outputId":"1d824d41-a4d8-4caa-a587-253aaa3b8b43"},"outputs":[{"output_type":"stream","name":"stdout","text":["🔍 Vérification des expériences déjà terminées dans 'OC Projet 8'...\n","✅ 34 expériences terminées trouvées:\n"," - FPN_efficientnetb0_finetune_augTrue_prod\n"," - FPN_efficientnetb0_finetune_augTrue_test\n"," - FPN_efficientnetb0_frozen_augFalse_prod\n"," - FPN_efficientnetb0_frozen_augFalse_test\n"," - FPN_efficientnetb0_frozen_augTrue_prod\n"," - FPN_efficientnetb0_frozen_augTrue_test\n"," - FPN_mobilenetv2_frozen_augFalse_prod\n"," - FPN_mobilenetv2_frozen_augFalse_test\n"," - FPN_mobilenetv2_frozen_augTrue_prod\n"," - FPN_mobilenetv2_frozen_augTrue_test\n"," - FPN_resnet34_finetune_augTrue_prod\n"," - FPN_resnet34_finetune_augTrue_test\n"," - FPN_resnet34_frozen_augFalse_prod\n"," - FPN_resnet34_frozen_augFalse_test\n"," - FPN_resnet34_frozen_augTrue_prod\n"," - FPN_resnet34_frozen_augTrue_test\n"," - FPN_vanilla_augFalse_prod\n"," - FPN_vanilla_augFalse_test\n"," - FPN_vanilla_augTrue_prod\n"," - FPN_vanilla_augTrue_test\n"," - U-Net_efficientnetb0_frozen_augFalse_prod\n"," - U-Net_efficientnetb0_frozen_augFalse_test\n"," - U-Net_efficientnetb0_frozen_augTrue_prod\n"," - U-Net_efficientnetb0_frozen_augTrue_test\n"," - U-Net_mobilenetv2_finetune_augTrue_prod\n"," - U-Net_mobilenetv2_finetune_augTrue_test\n"," - U-Net_mobilenetv2_frozen_augFalse_prod\n"," - U-Net_mobilenetv2_frozen_augFalse_test\n"," - U-Net_mobilenetv2_frozen_augTrue_prod\n"," - U-Net_mobilenetv2_frozen_augTrue_test\n"," - U-Net_vanilla_augFalse_prod\n"," - U-Net_vanilla_augFalse_test\n"," - U-Net_vanilla_augTrue_prod\n"," - U-Net_vanilla_augTrue_test\n","📊 Analyse de 17 configurations terminées\n"]},{"output_type":"display_data","data":{"text/plain":["Downloading artifacts: 0%| | 0/1 [00:00"],"image/png":"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\n"},"metadata":{}},{"output_type":"stream","name":"stdout","text":["📊 Diagrammes générés pour 17 expériences (IoU) et 17 expériences (Accuracy)\n","\n","✅ Analyse terminée. Utilisez show_detailed_results=True pour obtenir les données détaillées.\n"]}],"source":["compare_experiments_mlflow(experiment_name=mlflow_experiemnt)"]},{"cell_type":"markdown","metadata":{"id":"gqrw8X8wR5Q6"},"source":["### Analyse meilleur modèle"]},{"cell_type":"markdown","source":["Affichage des courbes d'apprentissage"],"metadata":{"id":"j9kVdhxVBA4S"}},{"cell_type":"code","source":["def display_training_curves_from_best_run(experiment_name=\"OC Projet 8\"):\n"," \"\"\"\n"," Trouve le meilleur modèle (selon test_mean_iou) et affiche son graphique de courbes d'entraînement.\n"," \"\"\"\n"," print(f\"\\n{'='*60}\")\n"," print(f\"📈 Affichage des courbes d'entraînement du meilleur modèle\")\n"," print(f\"🔬 Expérience MLflow: {experiment_name}\")\n"," print(f\"{'='*60}\")\n","\n"," client = mlflow.tracking.MlflowClient()\n","\n"," try:\n"," # Récupérer l'expérience par nom\n"," experiment = client.get_experiment_by_name(experiment_name)\n"," if experiment is None:\n"," print(f\"❌ Expérience '{experiment_name}' non trouvée.\")\n"," return\n","\n"," # Rechercher les runs terminés avec la métrique test_mean_iou\n"," runs_df = mlflow.search_runs(\n"," experiment_ids=[experiment.experiment_id],\n"," filter_string=\"status = 'FINISHED' and metrics.test_mean_iou >= 0\",\n"," order_by=[\"metrics.test_mean_iou DESC\"], # Trier par Mean IoU descendant\n"," max_results=1 # Prendre seulement le meilleur\n"," )\n","\n"," if runs_df.empty:\n"," print(\"❌ Aucun run terminé avec la métrique test_mean_iou trouvé.\")\n"," return\n","\n"," # Le meilleur run est le premier dans la liste triée\n"," best_run = runs_df.iloc[0]\n"," best_run_id = best_run['run_id']\n"," best_run_name = best_run.get('tags.mlflow.runName', 'Unknown')\n"," best_iou = best_run['metrics.test_mean_iou']\n","\n"," print(f\"🏆 Meilleur run trouvé:\")\n"," print(f\" • Run Name: {best_run_name}\")\n"," print(f\" • Run ID: {best_run_id}\")\n"," print(f\" • Test Mean IoU: {best_iou:.4f}\")\n","\n"," # Télécharger l'artefact 'training_curves.png'\n"," artifact_path = \"training_curves.png\"\n"," print(f\"\\n💾 Téléchargement de l'artefact '{artifact_path}'...\")\n","\n"," with tempfile.TemporaryDirectory() as tmpdir:\n"," local_path = mlflow.artifacts.download_artifacts(\n"," run_id=best_run_id,\n"," artifact_path=artifact_path,\n"," dst_path=tmpdir\n"," )\n"," print(f\"✅ Artefact téléchargé vers: {local_path}\")\n","\n"," # Afficher l'image\n"," print(\"\\n🖼️ Affichage de l'image...\")\n"," img = PILImage.open(local_path)\n"," plt.figure(figsize=(12, 8))\n"," plt.imshow(img)\n"," plt.title(f\"Courbes d'entraînement - {best_run_name}\", fontsize=14)\n"," plt.axis('off') # Désactiver les axes\n"," plt.show()\n","\n"," except Exception as e:\n"," print(f\"❌ Erreur lors de l l'affichage des courbes d'entraînement: {e}\")\n"," print(\"Vérifiez que l'artefact 'training_curves.png' existe pour le meilleur run.\")"],"metadata":{"id":"cQMW_Ri1dTE0","executionInfo":{"status":"ok","timestamp":1755008415674,"user_tz":-120,"elapsed":1,"user":{"displayName":"Grégoire Mureau","userId":"04855910922109670788"}}},"execution_count":27,"outputs":[]},{"cell_type":"code","source":["# Exécuter la fonction\n","display_training_curves_from_best_run(experiment_name=mlflow_experiemnt)"],"metadata":{"id":"K7FA2AlTg0X-","colab":{"base_uri":"https://localhost:8080/","height":959,"referenced_widgets":["1d6592ac880e4dcba15ef22d2775c89f","08db4adb6c7149a18ac6b5480b420c73","00f7a5c8476243198bc254d29eb84061","b44ce655454543139597898a77b6c7f3","bd6f7801ad924c5a84235741a900d0a6","2bca1e7773ad4677a594e69dbffb9288","d3fc0b7f9b594cd6b06dc0ba41dc9585","89ac9f115547479d90b493df7d117dd9","ead3b3fc71114acf8a8799b511748218","5e536cf5cc6e4b00bad4754f566b230f","926e00ef12bc4a5da492072bb677ba24"]},"executionInfo":{"status":"ok","timestamp":1755008419464,"user_tz":-120,"elapsed":3789,"user":{"displayName":"Grégoire Mureau","userId":"04855910922109670788"}},"outputId":"14a3d6bc-7439-4265-f69f-d7e877c614cb"},"execution_count":28,"outputs":[{"output_type":"stream","name":"stdout","text":["\n","============================================================\n","📈 Affichage des courbes d'entraînement du meilleur modèle\n","🔬 Expérience MLflow: OC Projet 8\n","============================================================\n","🏆 Meilleur run trouvé:\n"," • Run Name: FPN_efficientnetb0_finetune_augTrue_prod\n"," • Run ID: b6dd1ecdeb3f4b5c899b332f3c948cd7\n"," • Test Mean IoU: 0.7457\n","\n","💾 Téléchargement de l'artefact 'training_curves.png'...\n"]},{"output_type":"display_data","data":{"text/plain":["Downloading artifacts: 0%| | 0/1 [00:00"],"image/png":"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\n"},"metadata":{}}]},{"cell_type":"markdown","source":["Affichage de la matrice de confusion en pourcentages"],"metadata":{"id":"GFqCXkZEBMp2"}},{"cell_type":"code","execution_count":29,"metadata":{"id":"YNQDzY0BH-X0","executionInfo":{"status":"ok","timestamp":1755008419473,"user_tz":-120,"elapsed":8,"user":{"displayName":"Grégoire Mureau","userId":"04855910922109670788"}}},"outputs":[],"source":["def display_best_confusion_matrix(experiment_name=\"OC Projet 8\"):\n"," \"\"\"\n"," Trouve le modèle avec la meilleure valeur test_mean_iou et affiche sa matrice de confusion\n","\n"," Args:\n"," experiment_name (str): Nom de l'expérience MLflow\n"," \"\"\"\n"," import matplotlib.pyplot as plt\n"," import seaborn as sns\n"," import numpy as np\n"," import json\n"," import tempfile\n","\n"," # Récupérer les résultats des expériences\n"," completed_configs = get_completed_configs_info(experimental_configurations, experiment_name)\n","\n"," if not completed_configs:\n"," print(\"❌ Aucune expérience terminée trouvée\")\n"," return None\n","\n"," # Récupérer le mapping run_name -> run_id depuis MLflow\n"," experiment = mlflow.get_experiment_by_name(experiment_name)\n"," runs_df = mlflow.search_runs(\n"," experiment_ids=[experiment.experiment_id],\n"," filter_string=\"status = 'FINISHED'\",\n"," output_format=\"pandas\"\n"," )\n","\n"," run_name_to_id = {\n"," run.get('tags.mlflow.runName', ''): run['run_id']\n"," for _, run in runs_df.iterrows()\n"," }\n","\n"," print(f\"🔍 Recherche du meilleur modèle parmi {len(completed_configs)} configurations...\")\n","\n"," best_result = None\n"," best_iou = -1\n","\n"," # Trouver le meilleur modèle\n"," for config_idx, config, run_name in completed_configs:\n"," if run_name not in run_name_to_id:\n"," continue\n","\n"," run_id = run_name_to_id[run_name]\n","\n"," try:\n"," # Récupérer les métriques\n"," run = mlflow.get_run(run_id)\n"," metrics = run.data.metrics\n","\n"," test_mean_iou = metrics.get('test_mean_iou', None)\n","\n"," if test_mean_iou is not None and test_mean_iou > best_iou:\n"," best_iou = test_mean_iou\n"," best_result = {\n"," 'config_idx': config_idx,\n"," 'experiment': config['name'],\n"," 'architecture': config.get('architecture', 'Unknown'),\n"," 'encoder': config.get('encoder', 'None'),\n"," 'test_mean_iou': test_mean_iou,\n"," 'test_accuracy': metrics.get('test_sparse_categorical_accuracy', 'N/A'),\n"," 'run_name': run_name,\n"," 'run_id': run_id\n"," }\n","\n"," except Exception as e:\n"," print(f\"⚠️ Erreur lors de l'analyse de {run_name}: {e}\")\n"," continue\n","\n"," if best_result is None:\n"," print(\"❌ Aucun modèle avec test_mean_iou trouvé\")\n"," return None\n","\n"," print(f\"🏆 MEILLEUR MODÈLE TROUVÉ:\")\n"," print(f\" 📊 [{best_result['config_idx']}] {best_result['experiment']}\")\n"," print(f\" 🏗️ {best_result['architecture']} + {best_result['encoder']}\")\n"," print(f\" 🎯 Test Mean IoU: {best_result['test_mean_iou']:.4f}\")\n"," print(f\" 📈 Test Accuracy: {best_result['test_accuracy']:.4f}\")\n"," print(f\" 🆔 Run ID: {best_result['run_id']}\")\n"," print()\n","\n"," # Télécharger et charger la matrice de confusion\n"," try:\n"," with tempfile.TemporaryDirectory() as temp_dir:\n"," confusion_matrix_path = mlflow.artifacts.download_artifacts(\n"," run_id=best_result['run_id'],\n"," artifact_path=\"confusion_matrix.json\",\n"," dst_path=temp_dir\n"," )\n","\n"," with open(confusion_matrix_path, 'r') as f:\n"," confusion_data = json.load(f)\n","\n"," matrix = np.array(confusion_data['matrix'])\n"," class_names = confusion_data['class_names']\n"," accuracy = confusion_data.get('accuracy', 'N/A')\n"," total_samples = confusion_data.get('total_samples', 'N/A')\n","\n"," print(f\"📋 Informations sur la matrice de confusion:\")\n"," print(f\" 🔢 Nombre de classes: {len(class_names)}\")\n"," print(f\" 📊 Échantillons totaux: {total_samples:,}\" if total_samples != 'N/A' else \" 📊 Échantillons totaux: N/A\")\n"," print(f\" ✅ Précision globale: {accuracy:.4f}\" if accuracy != 'N/A' else \" ✅ Précision globale: N/A\")\n"," print()\n","\n"," # Calculer les métriques par classe\n"," print(\"📈 Métriques par classe:\")\n"," print(\"-\" * 50)\n","\n"," for i, class_name in enumerate(class_names):\n"," # Précision: TP / (TP + FP)\n"," tp = matrix[i, i]\n"," fp = np.sum(matrix[:, i]) - tp\n"," fn = np.sum(matrix[i, :]) - tp\n","\n"," precision = tp / (tp + fp) if (tp + fp) > 0 else 0\n"," recall = tp / (tp + fn) if (tp + fn) > 0 else 0\n"," f1 = 2 * (precision * recall) / (precision + recall) if (precision + recall) > 0 else 0\n","\n"," print(f\" {class_name:12s} - Précision: {precision:.4f} | Rappel: {recall:.4f} | F1: {f1:.4f}\")\n","\n"," print()\n","\n"," # Créer la visualisation\n"," plt.figure(figsize=(12, 10))\n","\n"," # Normaliser la matrice pour l'affichage (en pourcentages)\n"," matrix_normalized = matrix.astype('float') / matrix.sum(axis=1)[:, np.newaxis] * 100\n","\n"," # Créer le heatmap\n"," sns.heatmap(matrix_normalized,\n"," annot=True,\n"," fmt='.1f',\n"," cmap='Blues',\n"," xticklabels=class_names,\n"," yticklabels=class_names,\n"," cbar_kws={'label': 'Pourcentage (%)'})\n","\n"," plt.title(f'Matrice de Confusion - {best_result[\"experiment\"]}\\n'\n"," f'Mean IoU: {best_result[\"test_mean_iou\"]:.4f} | Accuracy: {accuracy:.4f}',\n"," fontsize=14, fontweight='bold', pad=20)\n"," plt.xlabel('Classes Prédites', fontsize=12)\n"," plt.ylabel('Classes Réelles', fontsize=12)\n"," plt.xticks(rotation=45, ha='right')\n"," plt.yticks(rotation=0)\n","\n"," plt.tight_layout()\n"," plt.show()\n","\n"," # Optionnel: afficher aussi la matrice brute\n"," print(\"🔢 Matrice de confusion brute (premiers/derniers éléments):\")\n"," print(\"=\" * 60)\n"," print(\"Classes:\", class_names)\n"," print(\"Matrice (format: lignes=réel, colonnes=prédit):\")\n"," print(matrix)\n","\n"," return {\n"," 'confusion_matrix': matrix,\n"," 'class_names': class_names,\n"," 'best_model_info': best_result,\n"," 'accuracy': accuracy,\n"," 'total_samples': total_samples\n"," }\n","\n"," except Exception as e:\n"," print(f\"❌ Erreur lors du téléchargement de la matrice de confusion: {e}\")\n"," print(f\" Run ID: {best_result['run_id']}\")\n"," print(\" Vérifiez que l'artefact 'confusion_matrix.json' existe pour ce run.\")\n"," return None"]},{"cell_type":"code","execution_count":30,"metadata":{"id":"ir7It9rTICQE","colab":{"base_uri":"https://localhost:8080/","height":1000,"referenced_widgets":["55ba2a5fc03e4a999966cdebb03209a7","3823431325244727b488ef7025d977f2","5f915fb81c68451d9d7d3c09a8336385","e4ca2e4089c348fda12aa2b853d9759f","4def49bb8249419ea6aa88e49c3ce325","5abaa586d30b406bbaeac2a71f009445","01b2dca58b8145c6aae6775fbd0babc9","70be4ff8242940ea9684ab9908d20bad","db448534ea3e4efa999437edf895edf3","f680f3899d764b0287e22b2bbf5af141","2655232ebc9449ba8b211600043739b9"]},"executionInfo":{"status":"ok","timestamp":1755008424127,"user_tz":-120,"elapsed":4653,"user":{"displayName":"Grégoire Mureau","userId":"04855910922109670788"}},"outputId":"7b3485b5-536d-4487-813c-5324526df80a"},"outputs":[{"output_type":"stream","name":"stdout","text":["🔍 Vérification des expériences déjà terminées dans 'OC Projet 8'...\n","✅ 34 expériences terminées trouvées:\n"," - FPN_efficientnetb0_finetune_augTrue_prod\n"," - FPN_efficientnetb0_finetune_augTrue_test\n"," - FPN_efficientnetb0_frozen_augFalse_prod\n"," - FPN_efficientnetb0_frozen_augFalse_test\n"," - FPN_efficientnetb0_frozen_augTrue_prod\n"," - FPN_efficientnetb0_frozen_augTrue_test\n"," - FPN_mobilenetv2_frozen_augFalse_prod\n"," - FPN_mobilenetv2_frozen_augFalse_test\n"," - FPN_mobilenetv2_frozen_augTrue_prod\n"," - FPN_mobilenetv2_frozen_augTrue_test\n"," - FPN_resnet34_finetune_augTrue_prod\n"," - FPN_resnet34_finetune_augTrue_test\n"," - FPN_resnet34_frozen_augFalse_prod\n"," - FPN_resnet34_frozen_augFalse_test\n"," - FPN_resnet34_frozen_augTrue_prod\n"," - FPN_resnet34_frozen_augTrue_test\n"," - FPN_vanilla_augFalse_prod\n"," - FPN_vanilla_augFalse_test\n"," - FPN_vanilla_augTrue_prod\n"," - FPN_vanilla_augTrue_test\n"," - U-Net_efficientnetb0_frozen_augFalse_prod\n"," - U-Net_efficientnetb0_frozen_augFalse_test\n"," - U-Net_efficientnetb0_frozen_augTrue_prod\n"," - U-Net_efficientnetb0_frozen_augTrue_test\n"," - U-Net_mobilenetv2_finetune_augTrue_prod\n"," - U-Net_mobilenetv2_finetune_augTrue_test\n"," - U-Net_mobilenetv2_frozen_augFalse_prod\n"," - U-Net_mobilenetv2_frozen_augFalse_test\n"," - U-Net_mobilenetv2_frozen_augTrue_prod\n"," - U-Net_mobilenetv2_frozen_augTrue_test\n"," - U-Net_vanilla_augFalse_prod\n"," - U-Net_vanilla_augFalse_test\n"," - U-Net_vanilla_augTrue_prod\n"," - U-Net_vanilla_augTrue_test\n","🔍 Recherche du meilleur modèle parmi 17 configurations...\n","🏆 MEILLEUR MODÈLE TROUVÉ:\n"," 📊 [15] FPN_efficientnetb0_finetune_augTrue\n"," 🏗️ FPN + efficientnetb0\n"," 🎯 Test Mean IoU: 0.7457\n"," 📈 Test Accuracy: 0.9123\n"," 🆔 Run ID: b6dd1ecdeb3f4b5c899b332f3c948cd7\n","\n"]},{"output_type":"display_data","data":{"text/plain":["Downloading artifacts: 0%| | 0/1 [00:00"],"image/png":"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\n"},"metadata":{}},{"output_type":"stream","name":"stdout","text":["🔢 Matrice de confusion brute (premiers/derniers éléments):\n","============================================================\n","Classes: ['flat', 'human', 'vehicle', 'construction', 'object', 'nature', 'sky', 'void']\n","Matrice (format: lignes=réel, colonnes=prédit):\n","[[9339912 11922 31996 27736 7639 19873 159 96403]\n"," [ 9298 240878 27954 32796 6215 7121 22 7213]\n"," [ 28923 20536 1630480 57085 7436 16871 280 13413]\n"," [ 37107 27385 39861 4762597 51086 145430 14571 61532]\n"," [ 12585 11363 14823 148802 240856 58198 2359 24774]\n"," [ 30020 6106 17487 161547 21272 3703362 10850 23316]\n"," [ 2 2 139 17203 1113 12668 701752 3902]\n"," [ 400815 21207 41511 245295 36253 56041 13049 2267598]]\n"]}],"source":["# Utilisation\n","confusion_result = display_best_confusion_matrix()"]},{"cell_type":"markdown","source":["Affichage de la répartition des modalités dans l'ensemble du dataset d'apprentissage (on passe par le json de matrice de confusion pour gagner du temps)"],"metadata":{"id":"KsVx1dEpBPph"}},{"cell_type":"code","source":["def analyze_class_distribution_from_confusion_matrix(experiment_name=\"OC Projet 8\",\n"," figsize=(14, 8),\n"," color_palette='viridis',\n"," show_values=True):\n"," \"\"\"\n"," Analyse la répartition des classes à partir de la matrice de confusion du meilleur modèle.\n"," Affiche un graphique en barres avec pourcentages (axe gauche) et nombres absolus (axe droite).\n","\n"," Args:\n"," experiment_name (str): Nom de l'expérience MLflow\n"," figsize (tuple): Taille de la figure (largeur, hauteur)\n"," color_palette (str): Palette de couleurs pour les barres\n"," show_values (bool): Afficher les valeurs sur les barres\n","\n"," Returns:\n"," dict: Dictionnaire contenant les statistiques de distribution\n"," \"\"\"\n"," import matplotlib.pyplot as plt\n"," import numpy as np\n"," import json\n"," import tempfile\n"," import seaborn as sns\n","\n"," # Récupérer les informations du meilleur modèle\n"," completed_configs = get_completed_configs_info(experimental_configurations, experiment_name)\n","\n"," if not completed_configs:\n"," print(\"❌ Aucune expérience terminée trouvée\")\n"," return None\n","\n"," # Récupérer le mapping run_name -> run_id depuis MLflow\n"," experiment = mlflow.get_experiment_by_name(experiment_name)\n"," runs_df = mlflow.search_runs(\n"," experiment_ids=[experiment.experiment_id],\n"," filter_string=\"status = 'FINISHED'\",\n"," output_format=\"pandas\"\n"," )\n","\n"," run_name_to_id = {\n"," run.get('tags.mlflow.runName', ''): run['run_id']\n"," for _, run in runs_df.iterrows()\n"," }\n","\n"," # Trouver le meilleur modèle (même logique que display_best_confusion_matrix)\n"," best_result = None\n"," best_iou = -1\n","\n"," for config_idx, config, run_name in completed_configs:\n"," if run_name not in run_name_to_id:\n"," continue\n","\n"," run_id = run_name_to_id[run_name]\n","\n"," try:\n"," run = mlflow.get_run(run_id)\n"," metrics = run.data.metrics\n"," test_mean_iou = metrics.get('test_mean_iou', None)\n","\n"," if test_mean_iou is not None and test_mean_iou > best_iou:\n"," best_iou = test_mean_iou\n"," best_result = {\n"," 'config_idx': config_idx,\n"," 'experiment': config['name'],\n"," 'architecture': config.get('architecture', 'Unknown'),\n"," 'encoder': config.get('encoder', 'None'),\n"," 'test_mean_iou': test_mean_iou,\n"," 'run_name': run_name,\n"," 'run_id': run_id\n"," }\n"," except Exception as e:\n"," continue\n","\n"," if best_result is None:\n"," print(\"❌ Aucun modèle avec test_mean_iou trouvé\")\n"," return None\n","\n"," print(f\"🏆 Analyse de distribution pour le meilleur modèle:\")\n"," print(f\" 📊 [{best_result['config_idx']}] {best_result['experiment']}\")\n"," print(f\" 🎯 Test Mean IoU: {best_result['test_mean_iou']:.4f}\")\n"," print()\n","\n"," # Télécharger et analyser la matrice de confusion\n"," try:\n"," with tempfile.TemporaryDirectory() as temp_dir:\n"," confusion_matrix_path = mlflow.artifacts.download_artifacts(\n"," run_id=best_result['run_id'],\n"," artifact_path=\"confusion_matrix.json\",\n"," dst_path=temp_dir\n"," )\n","\n"," with open(confusion_matrix_path, 'r') as f:\n"," confusion_data = json.load(f)\n","\n"," matrix = np.array(confusion_data['matrix'])\n"," class_names = confusion_data['class_names']\n","\n"," # Calculer la distribution des classes (somme par ligne = nombre total de pixels réels par classe)\n"," class_counts = np.sum(matrix, axis=1) # Somme par ligne pour obtenir le nombre total de chaque classe réelle\n"," total_pixels = np.sum(class_counts)\n"," class_percentages = (class_counts / total_pixels) * 100\n","\n"," # Statistiques détaillées\n"," print(\"📈 RÉPARTITION DES CLASSES DANS L'ÉCHANTILLON:\")\n"," print(\"=\" * 65)\n"," print(f\"{'Classe':<15} {'Nb Pixels':<12} {'Pourcentage':<12} {'Barre':<20}\")\n"," print(\"-\" * 65)\n","\n"," for i, class_name in enumerate(class_names):\n"," bar_length = int(class_percentages[i] / 2) # Barre visuelle\n"," bar = \"█\" * bar_length\n"," print(f\"{class_name:<15} {class_counts[i]:<12,} {class_percentages[i]:<11.2f}% {bar}\")\n","\n"," print(\"-\" * 65)\n"," print(f\"{'TOTAL':<15} {total_pixels:<12,} {'100.00%':<12}\")\n"," print()\n","\n"," # Créer le graphique\n"," fig, ax1 = plt.subplots(figsize=figsize)\n","\n"," # Utiliser une palette de couleurs\n"," colors = plt.cm.get_cmap(color_palette)(np.linspace(0, 1, len(class_names)))\n","\n"," # Barres avec les pourcentages (axe gauche)\n"," bars = ax1.bar(range(len(class_names)), class_percentages,\n"," color=colors, alpha=0.8, edgecolor='black', linewidth=1)\n","\n"," # Configuration axe gauche (pourcentages)\n"," ax1.set_xlabel('Classes', fontsize=12, fontweight='bold')\n"," ax1.set_ylabel('Pourcentage (%)', fontsize=12, fontweight='bold', color='blue')\n"," ax1.tick_params(axis='y', labelcolor='blue')\n"," ax1.set_xticks(range(len(class_names)))\n"," ax1.set_xticklabels(class_names, rotation=45, ha='right')\n"," ax1.grid(True, alpha=0.3, axis='y')\n","\n"," # Créer un second axe y pour les nombres absolus\n"," ax2 = ax1.twinx()\n"," ax2.set_ylabel('Nombre de Pixels', fontsize=12, fontweight='bold', color='red')\n"," ax2.tick_params(axis='y', labelcolor='red')\n","\n"," # Synchroniser les échelles\n"," ax2.set_ylim(0, np.max(class_counts) * 1.1)\n"," ax1.set_ylim(0, np.max(class_percentages) * 1.1)\n","\n"," # Ajouter les valeurs sur les barres si demandé\n"," if show_values:\n"," for i, (bar, count, pct) in enumerate(zip(bars, class_counts, class_percentages)):\n"," height = bar.get_height()\n"," # Pourcentage au-dessus de la barre\n"," ax1.text(bar.get_x() + bar.get_width()/2., height + 0.5,\n"," f'{pct:.1f}%', ha='center', va='bottom', fontweight='bold', fontsize=10)\n"," # Nombre absolu à l'intérieur de la barre\n"," ax1.text(bar.get_x() + bar.get_width()/2., height/2,\n"," f'{count:,}', ha='center', va='center',\n"," color='white', fontweight='bold', fontsize=9)\n","\n"," # Titre et mise en forme\n"," plt.title(f'Répartition des Classes par Pixels\\n'\n"," f'Total: {total_pixels:,} pixels | Mean IoU: {best_result[\"test_mean_iou\"]:.4f}',\n"," fontsize=14, fontweight='bold', pad=20)\n","\n"," plt.tight_layout()\n"," plt.show()\n","\n"," # Statistiques supplémentaires\n"," print(\"📊 STATISTIQUES COMPLÉMENTAIRES:\")\n"," print(\"-\" * 40)\n"," print(f\" 🔢 Nombre total de pixels: {total_pixels:,}\")\n"," print(f\" 📏 Nombre de classes: {len(class_names)}\")\n"," print(f\" 📈 Classe la plus représentée: {class_names[np.argmax(class_counts)]} ({np.max(class_percentages):.2f}%)\")\n"," print(f\" 📉 Classe la moins représentée: {class_names[np.argmin(class_counts)]} ({np.min(class_percentages):.2f}%)\")\n"," print(f\" ⚖️ Ratio max/min: {np.max(class_counts)/np.min(class_counts):.2f}\")\n","\n"," # Calcul de l'équilibrage des classes\n"," ideal_percentage = 100 / len(class_names)\n"," imbalance_scores = np.abs(class_percentages - ideal_percentage)\n"," avg_imbalance = np.mean(imbalance_scores)\n"," print(f\" ⚖️ Score de déséquilibre moyen: {avg_imbalance:.2f}% (0 = parfait équilibre)\")\n"," print()\n","\n"," return {\n"," 'class_names': class_names,\n"," 'class_counts': class_counts,\n"," 'class_percentages': class_percentages,\n"," 'total_pixels': total_pixels,\n"," 'best_model_info': best_result,\n"," 'most_common_class': class_names[np.argmax(class_counts)],\n"," 'least_common_class': class_names[np.argmin(class_counts)],\n"," 'imbalance_ratio': np.max(class_counts)/np.min(class_counts),\n"," 'avg_imbalance_score': avg_imbalance\n"," }\n","\n"," except Exception as e:\n"," print(f\"❌ Erreur lors de l'analyse de distribution: {e}\")\n"," print(f\" Run ID: {best_result['run_id']}\")\n"," return None"],"metadata":{"id":"LXs3Sr5pA6uJ","executionInfo":{"status":"ok","timestamp":1755008480321,"user_tz":-120,"elapsed":1,"user":{"displayName":"Grégoire Mureau","userId":"04855910922109670788"}}},"execution_count":39,"outputs":[]},{"cell_type":"code","source":["# Analyse simple du meilleur modèle\n","result = analyze_class_distribution_from_confusion_matrix(\"OC Projet 8\")\n","\n","# Avec personnalisation\n","result = analyze_class_distribution_from_confusion_matrix(\n"," experiment_name=\"OC Projet 8\",\n"," figsize=(16, 10),\n"," color_palette='plasma',\n"," show_values=True\n",")"],"metadata":{"id":"VoTFNfGhBY8x","colab":{"base_uri":"https://localhost:8080/","height":1000,"referenced_widgets":["1d541fd90d3648269b8117c9aca70f04","1e9c76b7e8ff40f4bedf9ad6c020bdcc","280ff39b4bde45699f8cc34c787c6fa6","6dda57d2132547d2b1cbd611d0001faa","74e14c97eb314d048e00e30068499873","3ca83ea8938d4fd382fac12d5217bc77","35d32db1f91547509be44cb317af0c18","c508bd4ce5a3429299d82e6e70c490a8","47d00c0e1aa44532bb874f757b609bee","2b6b2c92b68d4400bf0db19c7898cae7","2e4134947f0842e693fb0df1b5cd4ea1","54cf1a05e871423c89b83a34f7f0cf1f","35121abbca48482e95ead258bce73029","fbb125df3c2245e796deea4bbacd9fce","f76bc9f28cdc4d2b969124c2fd47f36b","56e123a65afc4cb5b20126c627d7b648","41dbcf9cfed64082aaaf8e791ffd8876","8541590a113a439f9f84ab89b80492ab","b7dce827c17d47488b0e8ba7dd5f881e","fa219a9a89a24b2b850e0e965139c5c0","89a5cc92bbbf464e939aae29180c4af5","916894e0fad845f98afaaea3042ded6c"]},"executionInfo":{"status":"ok","timestamp":1755008480314,"user_tz":-120,"elapsed":8133,"user":{"displayName":"Grégoire Mureau","userId":"04855910922109670788"}},"outputId":"0aa12419-91fd-45b1-843e-08ce44dcb528"},"execution_count":38,"outputs":[{"output_type":"stream","name":"stdout","text":["🔍 Vérification des expériences déjà terminées dans 'OC Projet 8'...\n","✅ 34 expériences terminées trouvées:\n"," - FPN_efficientnetb0_finetune_augTrue_prod\n"," - FPN_efficientnetb0_finetune_augTrue_test\n"," - FPN_efficientnetb0_frozen_augFalse_prod\n"," - FPN_efficientnetb0_frozen_augFalse_test\n"," - FPN_efficientnetb0_frozen_augTrue_prod\n"," - FPN_efficientnetb0_frozen_augTrue_test\n"," - FPN_mobilenetv2_frozen_augFalse_prod\n"," - FPN_mobilenetv2_frozen_augFalse_test\n"," - FPN_mobilenetv2_frozen_augTrue_prod\n"," - FPN_mobilenetv2_frozen_augTrue_test\n"," - FPN_resnet34_finetune_augTrue_prod\n"," - FPN_resnet34_finetune_augTrue_test\n"," - FPN_resnet34_frozen_augFalse_prod\n"," - FPN_resnet34_frozen_augFalse_test\n"," - FPN_resnet34_frozen_augTrue_prod\n"," - FPN_resnet34_frozen_augTrue_test\n"," - FPN_vanilla_augFalse_prod\n"," - FPN_vanilla_augFalse_test\n"," - FPN_vanilla_augTrue_prod\n"," - FPN_vanilla_augTrue_test\n"," - U-Net_efficientnetb0_frozen_augFalse_prod\n"," - U-Net_efficientnetb0_frozen_augFalse_test\n"," - U-Net_efficientnetb0_frozen_augTrue_prod\n"," - U-Net_efficientnetb0_frozen_augTrue_test\n"," - U-Net_mobilenetv2_finetune_augTrue_prod\n"," - U-Net_mobilenetv2_finetune_augTrue_test\n"," - U-Net_mobilenetv2_frozen_augFalse_prod\n"," - U-Net_mobilenetv2_frozen_augFalse_test\n"," - U-Net_mobilenetv2_frozen_augTrue_prod\n"," - U-Net_mobilenetv2_frozen_augTrue_test\n"," - U-Net_vanilla_augFalse_prod\n"," - U-Net_vanilla_augFalse_test\n"," - U-Net_vanilla_augTrue_prod\n"," - U-Net_vanilla_augTrue_test\n","🏆 Analyse de distribution pour le meilleur modèle:\n"," 📊 [15] FPN_efficientnetb0_finetune_augTrue\n"," 🎯 Test Mean IoU: 0.7457\n","\n"]},{"output_type":"display_data","data":{"text/plain":["Downloading artifacts: 0%| | 0/1 [00:00"],"image/png":"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\n"},"metadata":{}},{"output_type":"stream","name":"stdout","text":["📊 STATISTIQUES COMPLÉMENTAIRES:\n","----------------------------------------\n"," 🔢 Nombre total de pixels: 25,088,000\n"," 📏 Nombre de classes: 8\n"," 📈 Classe la plus représentée: flat (38.01%)\n"," 📉 Classe la moins représentée: human (1.32%)\n"," ⚖️ Ratio max/min: 28.77\n"," ⚖️ Score de déséquilibre moyen: 9.21% (0 = parfait équilibre)\n","\n","🔍 Vérification des expériences déjà terminées dans 'OC Projet 8'...\n","✅ 34 expériences terminées trouvées:\n"," - FPN_efficientnetb0_finetune_augTrue_prod\n"," - FPN_efficientnetb0_finetune_augTrue_test\n"," - FPN_efficientnetb0_frozen_augFalse_prod\n"," - FPN_efficientnetb0_frozen_augFalse_test\n"," - FPN_efficientnetb0_frozen_augTrue_prod\n"," - FPN_efficientnetb0_frozen_augTrue_test\n"," - FPN_mobilenetv2_frozen_augFalse_prod\n"," - FPN_mobilenetv2_frozen_augFalse_test\n"," - FPN_mobilenetv2_frozen_augTrue_prod\n"," - FPN_mobilenetv2_frozen_augTrue_test\n"," - FPN_resnet34_finetune_augTrue_prod\n"," - FPN_resnet34_finetune_augTrue_test\n"," - FPN_resnet34_frozen_augFalse_prod\n"," - FPN_resnet34_frozen_augFalse_test\n"," - FPN_resnet34_frozen_augTrue_prod\n"," - FPN_resnet34_frozen_augTrue_test\n"," - FPN_vanilla_augFalse_prod\n"," - FPN_vanilla_augFalse_test\n"," - FPN_vanilla_augTrue_prod\n"," - FPN_vanilla_augTrue_test\n"," - U-Net_efficientnetb0_frozen_augFalse_prod\n"," - U-Net_efficientnetb0_frozen_augFalse_test\n"," - U-Net_efficientnetb0_frozen_augTrue_prod\n"," - U-Net_efficientnetb0_frozen_augTrue_test\n"," - U-Net_mobilenetv2_finetune_augTrue_prod\n"," - U-Net_mobilenetv2_finetune_augTrue_test\n"," - U-Net_mobilenetv2_frozen_augFalse_prod\n"," - U-Net_mobilenetv2_frozen_augFalse_test\n"," - U-Net_mobilenetv2_frozen_augTrue_prod\n"," - U-Net_mobilenetv2_frozen_augTrue_test\n"," - U-Net_vanilla_augFalse_prod\n"," - U-Net_vanilla_augFalse_test\n"," - U-Net_vanilla_augTrue_prod\n"," - U-Net_vanilla_augTrue_test\n","🏆 Analyse de distribution pour le meilleur modèle:\n"," 📊 [15] FPN_efficientnetb0_finetune_augTrue\n"," 🎯 Test Mean IoU: 0.7457\n","\n"]},{"output_type":"display_data","data":{"text/plain":["Downloading artifacts: 0%| | 0/1 [00:00"],"image/png":"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\n"},"metadata":{}},{"output_type":"stream","name":"stdout","text":["📊 STATISTIQUES COMPLÉMENTAIRES:\n","----------------------------------------\n"," 🔢 Nombre total de pixels: 25,088,000\n"," 📏 Nombre de classes: 8\n"," 📈 Classe la plus représentée: flat (38.01%)\n"," 📉 Classe la moins représentée: human (1.32%)\n"," ⚖️ Ratio max/min: 28.77\n"," ⚖️ Score de déséquilibre moyen: 9.21% (0 = parfait équilibre)\n","\n"]}]},{"cell_type":"markdown","metadata":{"id":"8-vXPIYNSALV"},"source":["### Mise en production meilleur modèle"]},{"cell_type":"code","source":["def register_best_model(experiment_name=\"OC Projet 8\", model_name=\"StreetsSegmentation\",\n"," force=False, specific_run_id=None):\n"," \"\"\"\n"," Enregistre le meilleur modèle (selon test_mean_iou) OU un modèle spécifique dans le Registry MLflow.\n","\n"," Args:\n"," experiment_name (str): Nom de l'expérience MLflow\n"," model_name (str): Nom du modèle dans le Registry\n"," force (bool): Si True, force la mise en production même si le score est moins bon\n"," specific_run_id (str): Si fourni, enregistre ce run spécifique au lieu du meilleur\n"," \"\"\"\n"," print(f\"\\n{'='*60}\")\n"," if specific_run_id:\n"," print(f\"🎯 Enregistrement d'un modèle SPÉCIFIQUE dans le Registry MLflow\")\n"," print(f\"🆔 Run ID spécifique: {specific_run_id}\")\n"," else:\n"," print(f\"🏆 Enregistrement du MEILLEUR modèle dans le Registry MLflow\")\n"," print(f\"🔬 Expérience MLflow: {experiment_name}\")\n"," print(f\"📦 Nom du modèle Registry: {model_name}\")\n"," if force:\n"," print(f\"⚡ Mode FORCE activé - Mise en prod forcée\")\n"," print(f\"{'='*60}\")\n","\n"," # 1. Si un run_id spécifique est fourni, l'utiliser directement\n"," if specific_run_id:\n"," print(f\"🎯 Mode spécifique activé - utilisation du run {specific_run_id}\")\n","\n"," try:\n"," # Vérifier que le run existe et récupérer ses métriques\n"," run = mlflow.get_run(specific_run_id)\n"," metrics = run.data.metrics\n"," params = run.data.params\n","\n"," test_mean_iou = metrics.get('test_mean_iou', None)\n"," test_accuracy = metrics.get('test_sparse_categorical_accuracy', None)\n"," run_name = run.data.tags.get('mlflow.runName', 'Unknown')\n","\n"," if test_mean_iou is None:\n"," print(\"⚠️ Le run spécifique n'a pas de métrique test_mean_iou\")\n"," if not force:\n"," print(\"❌ Utilisez force=True pour forcer l'enregistrement sans métriques\")\n"," return None\n","\n"," selected_result = {\n"," 'run_id': specific_run_id,\n"," 'test_mean_iou': test_mean_iou or -1.0,\n"," 'test_accuracy': test_accuracy,\n"," 'experiment_name': run_name,\n"," 'selection_mode': 'specific'\n"," }\n","\n"," print(f\"✅ Run spécifique trouvé:\")\n"," print(f\" • Run Name: {run_name}\")\n"," print(f\" • Run ID: {specific_run_id}\")\n"," if test_mean_iou:\n"," print(f\" • Test Mean IoU: {test_mean_iou:.4f}\")\n"," if test_accuracy:\n"," print(f\" • Test Accuracy: {test_accuracy:.4f}\")\n","\n"," except Exception as e:\n"," print(f\"❌ Erreur lors de la récupération du run spécifique {specific_run_id}: {e}\")\n"," return None\n","\n"," else:\n"," # 2. Mode automatique : trouver le meilleur modèle (logique existante)\n"," print(\"🔍 Mode automatique - recherche du meilleur modèle...\")\n","\n"," # Récupérer les configurations terminées (réutilise la logique existante)\n"," completed_configs = get_completed_configs_info(experimental_configurations, experiment_name)\n","\n"," if not completed_configs:\n"," print(\"❌ Aucune expérience terminée trouvée\")\n"," return None\n","\n"," # Récupérer les métriques MLflow pour chaque run\n"," experiment = mlflow.get_experiment_by_name(experiment_name)\n"," runs_df = mlflow.search_runs(\n"," experiment_ids=[experiment.experiment_id],\n"," filter_string=\"status = 'FINISHED'\",\n"," output_format=\"pandas\"\n"," )\n","\n"," run_name_to_id = {\n"," run.get('tags.mlflow.runName', ''): run['run_id']\n"," for _, run in runs_df.iterrows()\n"," }\n","\n"," print(f\"📊 Analyse de {len(completed_configs)} configurations terminées\")\n","\n"," # Collecter les résultats avec métriques\n"," results = []\n"," for config_idx, config, run_name in completed_configs:\n"," if run_name not in run_name_to_id:\n"," continue\n","\n"," run_id = run_name_to_id[run_name]\n","\n"," try:\n"," # Récupérer les métriques directement depuis MLflow\n"," run = mlflow.get_run(run_id)\n"," metrics = run.data.metrics\n","\n"," test_mean_iou = metrics.get('test_mean_iou', None)\n","\n"," if test_mean_iou is not None:\n"," results.append({\n"," 'config_idx': config_idx,\n"," 'config': config,\n"," 'run_name': run_name,\n"," 'run_id': run_id,\n"," 'test_mean_iou': test_mean_iou,\n"," 'experiment_name': config['name']\n"," })\n"," print(f\" ✅ [{config_idx}] {config['name']}: IoU={test_mean_iou:.4f}\")\n"," else:\n"," print(f\" ⚠️ [{config_idx}] {config['name']}: Pas de test_mean_iou\")\n","\n"," except Exception as e:\n"," print(f\" ❌ [{config_idx}] {config['name']}: Erreur - {e}\")\n"," continue\n","\n"," if not results:\n"," print(\"❌ Aucune expérience avec test_mean_iou trouvée\")\n"," return None\n","\n"," # Trouver le meilleur modèle\n"," best_result = max(results, key=lambda x: x['test_mean_iou'])\n","\n"," selected_result = {\n"," 'run_id': best_result['run_id'],\n"," 'test_mean_iou': best_result['test_mean_iou'],\n"," 'experiment_name': best_result['experiment_name'],\n"," 'selection_mode': 'automatic'\n"," }\n","\n"," print(f\"🏆 Meilleur modèle identifié:\")\n"," print(f\" • Expérience: {best_result['experiment_name']}\")\n"," print(f\" • Run ID: {best_result['run_id']}\")\n"," print(f\" • Test Mean IoU: {best_result['test_mean_iou']:.4f}\")\n","\n"," # 3. Vérifier si le modèle existe dans le Registry (reste identique)\n"," client = mlflow.client.MlflowClient()\n"," current_best_iou = -1.0\n"," model_exists = False\n"," latest_version = None\n","\n"," try:\n"," try:\n"," registered_model = client.get_registered_model(model_name)\n"," model_exists = True\n"," print(f\"\\n📦 Modèle '{model_name}' trouvé dans le Registry\")\n"," except mlflow.exceptions.RestException:\n"," model_exists = False\n"," print(f\"\\n📦 Modèle '{model_name}' n'existe pas encore dans le Registry\")\n","\n"," if model_exists:\n"," model_versions = client.search_model_versions(f\"name='{model_name}'\")\n"," if model_versions:\n"," latest_version = max(model_versions, key=lambda v: int(v.version))\n"," current_run = mlflow.get_run(latest_version.run_id)\n"," current_best_iou = current_run.data.metrics.get('test_mean_iou', -1.0)\n","\n"," print(f\" • Version la plus récente: {latest_version.version}\")\n"," print(f\" • Stage: {latest_version.current_stage}\")\n"," print(f\" • Test Mean IoU: {current_best_iou:.4f}\")\n","\n"," except Exception as e:\n"," print(f\"⚠️ Erreur lors de la vérification du Registry: {e}\")\n","\n"," # 4. Logique de comparaison adaptée\n"," if specific_run_id:\n"," # En mode spécifique, on enregistre toujours (sauf si force=False et pas de métriques)\n"," should_register = True\n"," if selected_result['test_mean_iou'] == -1.0 and not force:\n"," should_register = False\n"," print(\"❌ Le run spécifique n'a pas de métriques et force=False\")\n"," else:\n"," # En mode automatique, comparaison habituelle\n"," should_register = selected_result['test_mean_iou'] > current_best_iou or force\n","\n"," # 5. Enregistrement (logique existante adaptée)\n"," if should_register:\n"," if specific_run_id:\n"," print(f\"\\n🎯 Enregistrement du modèle spécifique...\")\n"," if selected_result['test_mean_iou'] > current_best_iou:\n"," action_msg = f\"Le modèle spécifique ({selected_result['test_mean_iou']:.4f}) est meilleur que l'actuel ({current_best_iou:.4f})\"\n"," elif force:\n"," action_msg = f\"Enregistrement FORCÉ du modèle spécifique\"\n"," else:\n"," action_msg = f\"Enregistrement du modèle spécifique\"\n"," print(f\" {action_msg}\")\n"," else:\n"," if not model_exists:\n"," print(f\"\\n🆕 Création du modèle '{model_name}' dans le Registry...\")\n"," action_msg = \"Création et enregistrement du premier modèle\"\n"," elif force and selected_result['test_mean_iou'] <= current_best_iou:\n"," print(f\"\\n⚡ FORCE MODE: Enregistrement forcé du modèle\")\n"," print(f\" Nouveau: {selected_result['test_mean_iou']:.4f} vs Actuel: {current_best_iou:.4f}\")\n"," action_msg = \"Enregistrement forcé de la nouvelle version\"\n"," else:\n"," print(f\"\\n🚀 Le nouveau modèle ({selected_result['test_mean_iou']:.4f}) est meilleur que l'actuel ({current_best_iou:.4f})\")\n"," action_msg = \"Enregistrement de la nouvelle version\"\n","\n"," print(f\" {action_msg}...\")\n","\n"," try:\n"," # Si le modèle n'existe pas, le créer\n"," if not model_exists:\n"," client.create_registered_model(\n"," name=model_name,\n"," description=f\"Modèle de segmentation des rues - Performance: {selected_result['test_mean_iou']:.4f}\"\n"," )\n"," print(f\" ✅ Modèle '{model_name}' créé dans le Registry\")\n","\n"," # Vérifier si une version existe déjà pour ce run spécifique\n"," existing_versions = [v for v in client.search_model_versions(f\"name='{model_name}'\")\n"," if v.run_id == selected_result['run_id']]\n","\n"," if existing_versions:\n"," model_version = existing_versions[0]\n"," print(f\" ℹ️ Version existante trouvée pour ce run: {model_version.version}\")\n"," else:\n"," print(\" 📝 Enregistrement d'une nouvelle version...\")\n","\n"," # Chercher l'artefact du modèle\n"," artifacts = client.list_artifacts(selected_result['run_id'])\n"," model_artifact_path = None\n","\n"," for artifact in artifacts:\n"," if any(pattern in artifact.path.lower() for pattern in\n"," ['model', 'best_model', '.keras', '.h5', '.pkl']):\n"," model_artifact_path = artifact.path\n"," break\n","\n"," if model_artifact_path is None:\n"," print(\" ❌ Aucun artefact de modèle trouvé dans le run\")\n"," return None\n","\n"," print(f\" 📁 Artefact trouvé: {model_artifact_path}\")\n","\n"," model_version = client.create_model_version(\n"," name=model_name,\n"," source=f\"runs:/{selected_result['run_id']}/{model_artifact_path}\",\n"," run_id=selected_result['run_id'],\n"," description=f\"Model from {selected_result['experiment_name']} - IoU: {selected_result['test_mean_iou']:.4f} - Mode: {selected_result['selection_mode']}\"\n"," )\n","\n"," # Ajouter les tags\n"," print(\" 🏷️ Ajout des tags de métadonnées...\")\n"," tags_to_add = {\n"," \"source_run_id\": selected_result['run_id'],\n"," \"test_mean_iou\": str(selected_result['test_mean_iou']),\n"," \"model_name\": model_name,\n"," \"source_experiment\": selected_result['experiment_name'],\n"," \"selection_mode\": selected_result['selection_mode'],\n"," \"registration_mode\": \"forced\" if force and selected_result['test_mean_iou'] <= current_best_iou else \"automatic\"\n"," }\n","\n"," for tag_key, tag_value in tags_to_add.items():\n"," client.set_model_version_tag(\n"," name=model_name,\n"," version=model_version.version,\n"," key=tag_key,\n"," value=tag_value\n"," )\n","\n"," # Promouvoir en Production\n"," print(f\" 🚀 Promotion de la version {model_version.version} en Production...\")\n"," client.transition_model_version_stage(\n"," name=model_name,\n"," version=model_version.version,\n"," stage=\"Production\",\n"," archive_existing_versions=True\n"," )\n","\n"," print(f\"✅ Modèle enregistré avec succès !\")\n"," print(f\" • Nom: {model_name}\")\n"," print(f\" • Version: {model_version.version}\")\n"," print(f\" • Stage: Production\")\n"," print(f\" • Test Mean IoU: {selected_result['test_mean_iou']:.4f}\")\n"," print(f\" • Mode de sélection: {selected_result['selection_mode']}\")\n","\n"," return {\n"," 'model_name': model_name,\n"," 'version': model_version.version,\n"," 'run_id': selected_result['run_id'],\n"," 'test_mean_iou': selected_result['test_mean_iou'],\n"," 'selection_mode': selected_result['selection_mode'],\n"," 'action': 'created' if not model_exists else 'updated',\n"," 'tags': tags_to_add\n"," }\n","\n"," except Exception as e:\n"," print(f\"❌ Erreur lors de l'enregistrement: {e}\")\n"," return None\n"," else:\n"," print(f\"\\n➡️ Aucun enregistrement effectué\")\n"," if specific_run_id:\n"," print(\" Le run spécifique n'a pas de métriques suffisantes\")\n"," print(\" 💡 Utilisez force=True pour forcer l'enregistrement\")\n"," else:\n"," print(f\" La version actuelle ({current_best_iou:.4f}) est déjà optimale\")\n"," print(\" 💡 Utilisez force=True pour forcer la mise en production\")\n","\n"," return {\n"," 'model_name': model_name,\n"," 'action': 'no_update',\n"," 'reason': 'Insufficient metrics' if specific_run_id else 'Current version is optimal',\n"," 'suggestion': 'Use force=True to override'\n"," }"],"metadata":{"id":"FxkVT7_IE9fI","executionInfo":{"status":"ok","timestamp":1755008439899,"user_tz":-120,"elapsed":20,"user":{"displayName":"Grégoire Mureau","userId":"04855910922109670788"}}},"execution_count":33,"outputs":[]},{"cell_type":"code","execution_count":34,"metadata":{"id":"80mWsWg4Gi7Q","colab":{"base_uri":"https://localhost:8080/"},"executionInfo":{"status":"ok","timestamp":1755008443511,"user_tz":-120,"elapsed":3613,"user":{"displayName":"Grégoire Mureau","userId":"04855910922109670788"}},"outputId":"41e6f555-be2d-4040-d7c3-9386d5040017"},"outputs":[{"output_type":"stream","name":"stdout","text":["\n","============================================================\n","🏆 Enregistrement du MEILLEUR modèle dans le Registry MLflow\n","🔬 Expérience MLflow: OC Projet 8\n","📦 Nom du modèle Registry: StreetsSegmentation\n","============================================================\n","🔍 Mode automatique - recherche du meilleur modèle...\n","🔍 Vérification des expériences déjà terminées dans 'OC Projet 8'...\n","✅ 34 expériences terminées trouvées:\n"," - FPN_efficientnetb0_finetune_augTrue_prod\n"," - FPN_efficientnetb0_finetune_augTrue_test\n"," - FPN_efficientnetb0_frozen_augFalse_prod\n"," - FPN_efficientnetb0_frozen_augFalse_test\n"," - FPN_efficientnetb0_frozen_augTrue_prod\n"," - FPN_efficientnetb0_frozen_augTrue_test\n"," - FPN_mobilenetv2_frozen_augFalse_prod\n"," - FPN_mobilenetv2_frozen_augFalse_test\n"," - FPN_mobilenetv2_frozen_augTrue_prod\n"," - FPN_mobilenetv2_frozen_augTrue_test\n"," - FPN_resnet34_finetune_augTrue_prod\n"," - FPN_resnet34_finetune_augTrue_test\n"," - FPN_resnet34_frozen_augFalse_prod\n"," - FPN_resnet34_frozen_augFalse_test\n"," - FPN_resnet34_frozen_augTrue_prod\n"," - FPN_resnet34_frozen_augTrue_test\n"," - FPN_vanilla_augFalse_prod\n"," - FPN_vanilla_augFalse_test\n"," - FPN_vanilla_augTrue_prod\n"," - FPN_vanilla_augTrue_test\n"," - U-Net_efficientnetb0_frozen_augFalse_prod\n"," - U-Net_efficientnetb0_frozen_augFalse_test\n"," - U-Net_efficientnetb0_frozen_augTrue_prod\n"," - U-Net_efficientnetb0_frozen_augTrue_test\n"," - U-Net_mobilenetv2_finetune_augTrue_prod\n"," - U-Net_mobilenetv2_finetune_augTrue_test\n"," - U-Net_mobilenetv2_frozen_augFalse_prod\n"," - U-Net_mobilenetv2_frozen_augFalse_test\n"," - U-Net_mobilenetv2_frozen_augTrue_prod\n"," - U-Net_mobilenetv2_frozen_augTrue_test\n"," - U-Net_vanilla_augFalse_prod\n"," - U-Net_vanilla_augFalse_test\n"," - U-Net_vanilla_augTrue_prod\n"," - U-Net_vanilla_augTrue_test\n","📊 Analyse de 17 configurations terminées\n"," ✅ [0] U-Net_vanilla_augFalse: IoU=0.7058\n"," ✅ [1] U-Net_vanilla_augTrue: IoU=0.6986\n"," ✅ [2] FPN_vanilla_augFalse: IoU=0.7171\n"," ✅ [3] FPN_vanilla_augTrue: IoU=0.7148\n"," ✅ [4] U-Net_mobilenetv2_frozen_augFalse: IoU=0.6992\n"," ✅ [5] U-Net_mobilenetv2_frozen_augTrue: IoU=0.6933\n"," ✅ [6] U-Net_efficientnetb0_frozen_augFalse: IoU=0.7045\n"," ✅ [7] U-Net_efficientnetb0_frozen_augTrue: IoU=0.7088\n"," ✅ [8] FPN_mobilenetv2_frozen_augFalse: IoU=0.7025\n"," ✅ [9] FPN_mobilenetv2_frozen_augTrue: IoU=0.7082\n"," ✅ [10] FPN_efficientnetb0_frozen_augFalse: IoU=0.7232\n"," ✅ [11] FPN_efficientnetb0_frozen_augTrue: IoU=0.7214\n"," ✅ [12] FPN_resnet34_frozen_augFalse: IoU=0.6940\n"," ✅ [13] FPN_resnet34_frozen_augTrue: IoU=0.7017\n"," ✅ [14] U-Net_mobilenetv2_finetune_augTrue: IoU=0.7130\n"," ✅ [15] FPN_efficientnetb0_finetune_augTrue: IoU=0.7457\n"," ✅ [16] FPN_resnet34_finetune_augTrue: IoU=0.7181\n","🏆 Meilleur modèle identifié:\n"," • Expérience: FPN_efficientnetb0_finetune_augTrue\n"," • Run ID: b6dd1ecdeb3f4b5c899b332f3c948cd7\n"," • Test Mean IoU: 0.7457\n","\n","📦 Modèle 'StreetsSegmentation' trouvé dans le Registry\n"," • Version la plus récente: 2\n"," • Stage: Production\n"," • Test Mean IoU: 0.7457\n","\n","➡️ Aucun enregistrement effectué\n"," La version actuelle (0.7457) est déjà optimale\n"," 💡 Utilisez force=True pour forcer la mise en production\n"]},{"output_type":"execute_result","data":{"text/plain":["{'model_name': 'StreetsSegmentation',\n"," 'action': 'no_update',\n"," 'reason': 'Current version is optimal',\n"," 'suggestion': 'Use force=True to override'}"]},"metadata":{},"execution_count":34}],"source":["# Enregistrement automatique (seulement si meilleur)\n","register_best_model(experiment_name=\"OC Projet 8\", model_name=\"StreetsSegmentation\")"]},{"cell_type":"markdown","metadata":{"id":"m5gJ0G9SK2Pn"},"source":["## Fonction d'inférence et démonstration"]},{"cell_type":"markdown","source":["Les fonctions utiles sont toutes dans les imports en début de notebook"],"metadata":{"id":"IrCVqitDg5sk"}},{"cell_type":"code","execution_count":35,"metadata":{"id":"Az_I9YlrRevb","colab":{"base_uri":"https://localhost:8080/","height":176,"referenced_widgets":["2ae48a6a658d4a73907805a100996517","d383c5202b4c4bde976a95486de93745","90f88dbfcd004e7b9bbf7199dff4635e","2b413c4a996e4da4a33d7398dca34b42","85b3ff6fb3e74a3098b44bc012adbe87","7b31a5d8152b442a988a4632744042c7","a5ec8267d2fe4adc9eff959a271d017e","9ddd627c1cb94ba0903dd844c177e8d9","1e6ae9080ec748dfab98f2f6d10b67cd","f21e5d87afc949bcbb55d2d41f104056","b2f12a682daf49dbaca240b262edb0dd"]},"executionInfo":{"status":"ok","timestamp":1755008462897,"user_tz":-120,"elapsed":19383,"user":{"displayName":"Grégoire Mureau","userId":"04855910922109670788"}},"outputId":"d7092efa-a84b-4039-90d7-7a9e70753620"},"outputs":[{"output_type":"stream","name":"stdout","text":["📦 Téléchargement du modèle StreetsSegmentation (run: b6dd1ecdeb3f4b5c899b332f3c948cd7)\n"]},{"output_type":"display_data","data":{"text/plain":["Downloading artifacts: 0%| | 0/1 [00:00"],"image/png":"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\n"},"metadata":{}},{"output_type":"stream","name":"stdout","text":["\n","📊 Statistiques du masque prédit:\n"," flat: 6853 pixels (13.7%)\n"," human: 7970 pixels (15.9%)\n"," vehicle: 9227 pixels (18.4%)\n"," construction: 19823 pixels (39.5%)\n"," object: 1727 pixels (3.4%)\n"," nature: 889 pixels (1.8%)\n"," sky: 782 pixels (1.6%)\n"," void: 2905 pixels (5.8%)\n"]}]},{"cell_type":"code","source":["# === EXEMPLE D'UTILISATION ===\n","\n","# 1. Charger la configuration des classes\n","mapping_config = load_cityscapes_config(\"../cityscapes_config.json\")\n","\n","# 2. Faire l'inférence sur une nouvelle image\n","predicted_mask1 = run_inference_and_visualize(\n"," image_path=\"./content/data/test_images_sample/original/frankfurt_000000_000294_leftImg8bit.png\",\n"," mask_path=\"./content/data/test_images_sample/mask/frankfurt_000000_000294_gtFine_labelIds.png\",\n"," model=model,\n"," encoder_name=encoder_name,\n"," img_size=script_img_size,\n"," mapping_config=mapping_config\n",")"],"metadata":{"id":"RqYbA_qD52V2","colab":{"base_uri":"https://localhost:8080/","height":1000},"executionInfo":{"status":"ok","timestamp":1755008472178,"user_tz":-120,"elapsed":4514,"user":{"displayName":"Grégoire Mureau","userId":"04855910922109670788"}},"outputId":"dab91a16-adb8-4422-c517-82ba2f58079f"},"execution_count":37,"outputs":[{"output_type":"stream","name":"stdout","text":["== Groupes de classes Cityscapes ==\n"," - flat : ['road', 'sidewalk', 'parking', 'rail track']\n"," - human : ['person', 'rider']\n"," - vehicle : ['car', 'truck', 'bus', 'train', 'motorcycle', 'bicycle', 'caravan', 'trailer']\n"," - construction : ['building', 'wall', 'fence', 'guard rail', 'bridge', 'tunnel']\n"," - object : ['pole', 'pole group', 'traffic sign', 'traffic light']\n"," - nature : ['vegetation', 'terrain']\n"," - sky : ['sky']\n"," - void : ['unlabeled', 'ego vehicle', 'rectification border', 'out of roi', 'static', 'dynamic', 'ground']\n","\n","== Mapping ID Cityscapes → Groupe ==\n"," ID 0 ('unlabeled') → groupe 'void' (7)\n"," ID 1 ('ego vehicle') → groupe 'void' (7)\n"," ID 2 ('rectification border') → groupe 'void' (7)\n"," ID 3 ('out of roi') → groupe 'void' (7)\n"," ID 4 ('static') → groupe 'void' (7)\n"," ID 5 ('dynamic') → groupe 'void' (7)\n"," ID 6 ('ground') → groupe 'void' (7)\n"," ID 7 ('road') → groupe 'flat' (0)\n"," ID 8 ('sidewalk') → groupe 'flat' (0)\n"," ID 9 ('parking') → groupe 'flat' (0)\n"," ID 10 ('rail track') → groupe 'flat' (0)\n"," ID 11 ('building') → groupe 'construction' (3)\n"," ID 12 ('wall') → groupe 'construction' (3)\n"," ID 13 ('fence') → groupe 'construction' (3)\n"," ID 14 ('guard rail') → groupe 'construction' (3)\n"," ID 15 ('bridge') → groupe 'construction' (3)\n"," ID 16 ('tunnel') → groupe 'construction' (3)\n"," ID 17 ('pole') → groupe 'object' (4)\n"," ID 18 ('pole group') → groupe 'object' (4)\n"," ID 19 ('traffic light') → groupe 'object' (4)\n"," ID 20 ('traffic sign') → groupe 'object' (4)\n"," ID 21 ('vegetation') → groupe 'nature' (5)\n"," ID 22 ('terrain') → groupe 'nature' (5)\n"," ID 23 ('sky') → groupe 'sky' (6)\n"," ID 24 ('person') → groupe 'human' (1)\n"," ID 25 ('rider') → groupe 'human' (1)\n"," ID 26 ('car') → groupe 'vehicle' (2)\n"," ID 27 ('truck') → groupe 'vehicle' (2)\n"," ID 28 ('bus') → groupe 'vehicle' (2)\n"," ID 29 ('caravan') → groupe 'vehicle' (2)\n"," ID 30 ('trailer') → groupe 'vehicle' (2)\n"," ID 31 ('train') → groupe 'vehicle' (2)\n"," ID 32 ('motorcycle') → groupe 'vehicle' (2)\n"," ID 33 ('bicycle') → groupe 'vehicle' (2)\n","\n","== Liste des groupes et couleurs associées ==\n"," Groupe 0: flat - Couleur RGB : [128 64 128]\n"," Groupe 1: human - Couleur RGB : [220 20 60]\n"," Groupe 2: vehicle - Couleur RGB : [ 0 0 142]\n"," Groupe 3: construction - Couleur RGB : [70 70 70]\n"," Groupe 4: object - Couleur RGB : [220 220 0]\n"," Groupe 5: nature - Couleur RGB : [107 142 35]\n"," Groupe 6: sky - Couleur RGB : [ 70 130 180]\n"," Groupe 7: void - Couleur RGB : [0 0 0]\n","\n","Nombre total de groupes/classes : 8\n","=== INFÉRENCE SUR frankfurt_000000_000294_leftImg8bit.png ===\n","🔄 Préprocessing de l'image: frankfurt_000000_000294_leftImg8bit.png\n","ℹ️ Modèle vanilla - pas de preprocessing\n","🔄 Préprocessing du masque: frankfurt_000000_000294_gtFine_labelIds.png\n"," Valeurs uniques avant mapping: [ 3 2 11 23 17 4 20 21 26 24]...\n"," Valeurs uniques après mapping: [7 3 6 4 5 2 1 0]\n","🔮 Exécution de l'inférence...\n","✅ Inférence terminée - Shape: (224, 224)\n"]},{"output_type":"display_data","data":{"text/plain":["