OpenImagesDataLoader<T>

Models & Types · AiDotNet.Data.Vision.Benchmarks

Loads the Open Images V7 object detection dataset (~9M images, 600 categories).

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

How It Works

Open Images expects:

The CSV annotation format contains: ImageID, Source, LabelName, Confidence, XMin, XMax, YMin, YMax. Bounding box coordinates are already normalized to [0, 1]. Labels are stored as Tensor[N, MaxDetections, 5] (class_index, xmin, ymin, width, height).

Constructors

Constructor Summary
OpenImagesDataLoader(OpenImagesDataLoaderOptions options) Creates a new Open Images 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)