BigEarthNetDataLoader<T>
Models & Types · AiDotNet.Data.Vision.Benchmarks
Loads the BigEarthNet multi-label remote sensing dataset (590K Sentinel-2 patches, 19 or 43 classes).
Inheritance: DataLoaderBase<T> → InputOutputDataLoaderBase<T, Tensor<T>, Tensor<T>> → BigEarthNetDataLoader<T>
How It Works
BigEarthNet expects:
Each patch directory contains GeoTIFF band files and a JSON metadata file with CORINE Land Cover labels. Labels are multi-hot encoded as Tensor[N, NumClasses]. For RGB mode (NumBands=3), only B04(R), B03(G), B02(B) are loaded.
Constructors
| Constructor | Summary |
|---|---|
BigEarthNetDataLoader(BigEarthNetDataLoaderOptions options) |
Creates a new BigEarthNet data loader. |
Properties
| Property | Summary |
|---|---|
int FeatureCount { get; } |
|
int OutputDimension { get; } |
|
int TotalCount { get; } |
|
string Description { get; } |
|
string Name { get; } |
Methods
| Method | Summary |
|---|---|
ValueTuple<IInputOutputDataLoader<T, Tensor<T>, Tensor<T>>, IInputOutputDataLoader<T, Tensor<T>, Tensor<T>>, IInputOutputDataLoader<T, Tensor<T>, Tensor<T>>> Split(double trainRatio, double validationRatio, int? seed) |