Stl10DataLoaderOptions
Options & Configuration · AiDotNet.Data.Vision.Benchmarks
Configuration options for the STL-10 image classification dataset (Coates et al. 2011).
How It Works
STL-10 is a 10-class, 96×96 RGB image classification benchmark with 500 labeled train + 800 test images per class, plus 100,000 unlabeled images for self-supervised pretraining. Used widely for SSL/pretraining studies since the unlabeled split is large compared to the small labeled split.
Constructors
| Constructor | Summary |
|---|---|
Stl10DataLoaderOptions() |
Properties
| Property | Summary |
|---|---|
DatasetSplit Split { get; set; } |
Gets or sets which split of the dataset is loaded. |
bool AutoDownload { get; set; } |
Gets or sets whether the dataset is downloaded into the data directory when it is not there. |
bool Normalize { get; set; } |
Normalize byte pixel values to [0, 1] when true (default), or keep raw 0..255 when false. |
bool UseUnlabeled { get; set; } |
When true, returns the unlabeled split (100k images) instead of train/test. |
int? MaxSamples { get; set; } |
Gets or sets the most samples loaded; null loads the whole split. |
string DataPath { get; set; } |
Gets or sets the dataset root directory; null uses the default data directory for this dataset. |
Methods
| Method | Summary |
|---|---|
void Validate() |
Validates that all option values are within acceptable ranges. |