GlueDataLoader<T>
Models & Types · AiDotNet.Data.Text.Benchmarks
Loads GLUE benchmark sub-tasks (CoLA, SST-2, MRPC, QQP, STS-B, MNLI, QNLI, RTE, WNLI).
Inheritance: DataLoaderBase<T> → InputOutputDataLoaderBase<T, Tensor<T>, Tensor<T>> → GlueDataLoader<T>
How It Works
GLUE expects TSV files:
Features are token-index encoded text Tensor[N, MaxSequenceLength]. Labels are one-hot Tensor[N, NumClasses] (binary for most tasks, 3-class for MNLI, regression for STS-B).
Constructors
| Constructor | Summary |
|---|---|
GlueDataLoader(GlueDataLoaderOptions options) |
Creates a new GLUE 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) |