CocoDetectionDataLoader<T>

Models & Types · AiDotNet.Data.Vision.Benchmarks

Loads the COCO 2017 object detection dataset (118K train / 5K val, 80 categories).

Inheritance: DataLoaderBase<T> → InputOutputDataLoaderBase<T, Tensor<T>, Tensor<T>> → CocoDetectionDataLoader<T>

How It Works

COCO Detection expects:

Labels are stored as Tensor[N, MaxDetections, 5] where each detection is (class_id, x, y, w, h) with bounding box coordinates normalized to [0, 1]. Unused detection slots are zero-padded.

Constructors

Constructor Summary
CocoDetectionDataLoader(CocoDetectionDataLoaderOptions options) Creates a new COCO Detection 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)