CheXpertDataLoader<T>

Models & Types · AiDotNet.Data.Vision.Benchmarks

Loads the CheXpert chest radiograph dataset (224K images, 14 observations with uncertainty).

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

How It Works

CheXpert expects:

The CSV has columns: Path, Sex, Age, Frontal/Lateral, AP/PA, then 14 observation columns. Observation values: 1.0 (positive), 0.0 (negative), -1.0 (uncertain), blank (not mentioned). Labels are stored as Tensor[N, 14] with uncertainty handled per the UncertaintyPolicy option.

Constructors

Constructor Summary
CheXpertDataLoader(CheXpertDataLoaderOptions options) Creates a new CheXpert data loader.

Properties

Property Summary
IReadOnlyList<string> ObservationNames { get; } Gets the observation label names.
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)