Welcome to detectools’s documentation!

Detectools in an overlay of PyTorch for object detection and instances segmentation tasks in images. It’s development is supported by the PHENOME - EMPHASIS network.

Contents:

• Main Features
  • Loading data
  • Trainning process
  • Inference process
• Basics
  • set_lib_mode()
  • Task
• Usefull features
  • Visualisation
• Models
  • Mask2Former
  • YoloDetection
  • Yolov8Segmentation
• Formats
  • Annotation
  • BatchedFormats
  • DetectionAnnotation
  • DetectionFormat
  • Format
  • SegmentationAnnotation
  • SegmentationFormat
• Metrics
  • ClassifF1score
  • ClassifPrecision
  • ClassifRecall
  • DetectF1score
  • DetectPrecision
  • DetectRecall
  • MeanAP
  • SemanticAccuracy
  • SemanticF1score
  • SemanticIoU
• Utils
  • HiddenPrints
  • load_json()
  • raw_cocodict()
  • write_json()
• Going deeper
  • InferenceImage
  • BaseModel
  • BaseAnnotation
  • BaseFormat
  • DetectMetric
  • ClassifMetric
  • SemanticSegmentationMetric
  • MetricCompose
• Functionnals
  • Metrics
  • Inference
• Detection tutorial
  • Data structure
  • 0. Setup envrionnement for tutorial
  • 1. Data preparation
  • 2. Data format
  • 3. DetectionDataset & DetectionLoader
  • 4. Augmentation
  • 5. Trainning
  • Inference
• Instance segmentation tutorial
  • Data structure
  • 0. Setup envrionnement for tutorial
  • 1. Data preparation
  • 2. Data format
  • 3. DetectionDataset & DetectionLoader
  • 4. Augmentation
  • 5. Trainning
  • Inference
• Contribute
• Known issues
  • Patchification

Indices and tables

• Index

• Module Index

• Search Page