Module burn::data::dataset::vision

Vision datasets.

Structs

• BoundingBox
  Object detection bounding box annotation.
• ImageDatasetItem
  Image dataset item.
• ImageFolderDataset
  A generic dataset to load images from disk.
• MnistDataset
  The MNIST dataset consists of 70,000 28x28 black-and-white images in 10 classes (one for each digits), with 7,000 images per class. There are 60,000 training images and 10,000 test images.
• MnistItem
  MNIST item.
• SegmentationMask
  Segmentation mask annotation. For semantic segmentation, a mask has a single channel (C = 1). For instance segmentation, there may be multiple masks per image (C >= 1).

Enums

• Annotation
  Annotation type for different tasks.
• ImageLoaderError
  Error type for ImageFolderDataset.
• PixelDepth
  Image data type.