Agrupamento de modelos
📚 This guide explains how to use YOLOv5 🚀 model ensembling during testing and inference for improved mAP and Recall.
De https://en.wikipedia.org/wiki/Ensemble_learning:
Ensemble modeling is a process where multiple diverse models are created to predict an outcome, either by using many different modeling algorithms or using different training data sets. The ensemble model then aggregates the prediction of each base model and results in once final prediction for the unseen data. The motivation for using ensemble models is to reduce the generalization error of the prediction. As long as the base models are diverse and independent, the prediction error of the model decreases when the ensemble approach is used. The approach seeks the wisdom of crowds in making a prediction. Even though the ensemble model has multiple base models within the model, it acts and performs as a single model.
Antes de começares
Clona o repo e instala o requirements.txt em um Python>=3.8.0 incluindo PyTorch>=1.8. Os modelos e conjuntos de dados são descarregados automaticamente a partir daversão mais recente de YOLOv5 .
git clone https://github.com/ultralytics/yolov5 # clone
cd yolov5
pip install -r requirements.txt # install
Testa normalmente
Antes de fazer o ensembling, queremos estabelecer o desempenho de base de um único modelo. Este comando testa o YOLOv5x no COCO val2017 com um tamanho de imagem de 640 pixels. yolov5x.pt é o maior e mais preciso modelo disponível. Outras opções são yolov5s.pt, yolov5m.pt e yolov5l.ptou o teu próprio ponto de controlo a partir do treino de um conjunto de dados personalizado ./weights/best.pt. Para mais informações sobre todos os modelos disponíveis, consulta o nosso README tabela.
Saída:
val: data=./data/coco.yaml, weights=['yolov5x.pt'], batch_size=32, imgsz=640, conf_thres=0.001, iou_thres=0.65, task=val, device=, single_cls=False, augment=False, verbose=False, save_txt=False, save_hybrid=False, save_conf=False, save_json=True, project=runs/val, name=exp, exist_ok=False, half=True
YOLOv5 🚀 v5.0-267-g6a3ee7c torch 1.9.0+cu102 CUDA:0 (Tesla P100-PCIE-16GB, 16280.875MB)
Fusing layers...
Model Summary: 476 layers, 87730285 parameters, 0 gradients
val: Scanning '../datasets/coco/val2017' images and labels...4952 found, 48 missing, 0 empty, 0 corrupted: 100% 5000/5000 [00:01<00:00, 2846.03it/s]
val: New cache created: ../datasets/coco/val2017.cache
Class Images Labels P R mAP@.5 mAP@.5:.95: 100% 157/157 [02:30<00:00, 1.05it/s]
all 5000 36335 0.746 0.626 0.68 0.49
Speed: 0.1ms pre-process, 22.4ms inference, 1.4ms NMS per image at shape (32, 3, 640, 640) # <--- baseline speed
Evaluating pycocotools mAP... saving runs/val/exp/yolov5x_predictions.json...
...
Average Precision (AP) @[ IoU=0.50:0.95 | area= all | maxDets=100 ] = 0.504 # <--- baseline mAP
Average Precision (AP) @[ IoU=0.50 | area= all | maxDets=100 ] = 0.688
Average Precision (AP) @[ IoU=0.75 | area= all | maxDets=100 ] = 0.546
Average Precision (AP) @[ IoU=0.50:0.95 | area= small | maxDets=100 ] = 0.351
Average Precision (AP) @[ IoU=0.50:0.95 | area=medium | maxDets=100 ] = 0.551
Average Precision (AP) @[ IoU=0.50:0.95 | area= large | maxDets=100 ] = 0.644
Average Recall (AR) @[ IoU=0.50:0.95 | area= all | maxDets= 1 ] = 0.382
Average Recall (AR) @[ IoU=0.50:0.95 | area= all | maxDets= 10 ] = 0.628
Average Recall (AR) @[ IoU=0.50:0.95 | area= all | maxDets=100 ] = 0.681 # <--- baseline mAR
Average Recall (AR) @[ IoU=0.50:0.95 | area= small | maxDets=100 ] = 0.524
Average Recall (AR) @[ IoU=0.50:0.95 | area=medium | maxDets=100 ] = 0.735
Average Recall (AR) @[ IoU=0.50:0.95 | area= large | maxDets=100 ] = 0.826
Teste de conjunto
Vários modelos pré-treinados podem ser agrupados no momento do teste e da inferência, bastando acrescentar modelos adicionais ao --weights em qualquer comando val.py ou detect.py existente. Este exemplo testa um conjunto de 2 modelos juntos:
- YOLOv5x
- YOLOv5l6
Saída:
val: data=./data/coco.yaml, weights=['yolov5x.pt', 'yolov5l6.pt'], batch_size=32, imgsz=640, conf_thres=0.001, iou_thres=0.6, task=val, device=, single_cls=False, augment=False, verbose=False, save_txt=False, save_hybrid=False, save_conf=False, save_json=True, project=runs/val, name=exp, exist_ok=False, half=True
YOLOv5 🚀 v5.0-267-g6a3ee7c torch 1.9.0+cu102 CUDA:0 (Tesla P100-PCIE-16GB, 16280.875MB)
Fusing layers...
Model Summary: 476 layers, 87730285 parameters, 0 gradients # Model 1
Fusing layers...
Model Summary: 501 layers, 77218620 parameters, 0 gradients # Model 2
Ensemble created with ['yolov5x.pt', 'yolov5l6.pt'] # Ensemble notice
val: Scanning '../datasets/coco/val2017.cache' images and labels... 4952 found, 48 missing, 0 empty, 0 corrupted: 100% 5000/5000 [00:00<00:00, 49695545.02it/s]
Class Images Labels P R mAP@.5 mAP@.5:.95: 100% 157/157 [03:58<00:00, 1.52s/it]
all 5000 36335 0.747 0.637 0.692 0.502
Speed: 0.1ms pre-process, 39.5ms inference, 2.0ms NMS per image at shape (32, 3, 640, 640) # <--- ensemble speed
Evaluating pycocotools mAP... saving runs/val/exp3/yolov5x_predictions.json...
...
Average Precision (AP) @[ IoU=0.50:0.95 | area= all | maxDets=100 ] = 0.515 # <--- ensemble mAP
Average Precision (AP) @[ IoU=0.50 | area= all | maxDets=100 ] = 0.699
Average Precision (AP) @[ IoU=0.75 | area= all | maxDets=100 ] = 0.557
Average Precision (AP) @[ IoU=0.50:0.95 | area= small | maxDets=100 ] = 0.356
Average Precision (AP) @[ IoU=0.50:0.95 | area=medium | maxDets=100 ] = 0.563
Average Precision (AP) @[ IoU=0.50:0.95 | area= large | maxDets=100 ] = 0.668
Average Recall (AR) @[ IoU=0.50:0.95 | area= all | maxDets= 1 ] = 0.387
Average Recall (AR) @[ IoU=0.50:0.95 | area= all | maxDets= 10 ] = 0.638
Average Recall (AR) @[ IoU=0.50:0.95 | area= all | maxDets=100 ] = 0.689 # <--- ensemble mAR
Average Recall (AR) @[ IoU=0.50:0.95 | area= small | maxDets=100 ] = 0.526
Average Recall (AR) @[ IoU=0.50:0.95 | area=medium | maxDets=100 ] = 0.743
Average Recall (AR) @[ IoU=0.50:0.95 | area= large | maxDets=100 ] = 0.844
Inferência de conjunto
Anexa modelos extra ao --weights argumento para executar a inferência de conjunto:
Saída:
YOLOv5 🚀 v5.0-267-g6a3ee7c torch 1.9.0+cu102 CUDA:0 (Tesla P100-PCIE-16GB, 16280.875MB)
Fusing layers...
Model Summary: 476 layers, 87730285 parameters, 0 gradients
Fusing layers...
Model Summary: 501 layers, 77218620 parameters, 0 gradients
Ensemble created with ['yolov5x.pt', 'yolov5l6.pt']
image 1/2 /content/yolov5/data/images/bus.jpg: 640x512 4 persons, 1 bus, 1 tie, Done. (0.063s)
image 2/2 /content/yolov5/data/images/zidane.jpg: 384x640 3 persons, 2 ties, Done. (0.056s)
Results saved to runs/detect/exp2
Done. (0.223s)
Ambientes suportados
Ultralytics fornece uma gama de ambientes prontos a utilizar, cada um pré-instalado com dependências essenciais, tais como CUDA, CUDNN, Python, e PyTorchpara dar início aos teus projectos.
- Grátis GPU Notebooks:
- Google Nuvem: Guia de início rápido do GCP
- Amazon: Guia de início rápido do AWS
- Azure: Guia de início rápido do AzureML
- Docker: Guia de início rápido do Docker
Estado do projeto
Este emblema indica que todos os testes de Integração Contínua (CI) do YOLOv5 GitHub Actions foram aprovados com êxito. Esses testes de CI verificam rigorosamente a funcionalidade e o desempenho do YOLOv5 em vários aspectos principais: treinamento, validação, inferência, exportação e benchmarks. Eles garantem uma operação consistente e confiável no macOS, Windows e Ubuntu, com testes realizados a cada 24 horas e a cada novo commit.
