MbppDataLoaderOptions
Options & Configuration · AiDotNet.Data.Text.Benchmarks
Configuration options for the Mostly Basic Python Problems (MBPP) loader (Austin et al. 2021).
How It Works
MBPP is a 1,000-problem benchmark of basic Python programming tasks, each with a natural-language description, canonical reference solution, and 3 unit tests. Standard split: 974 train / 90 val / 500 test (with 100 reserved for prompt). Used as the entry-level code-generation benchmark alongside HumanEval.
Constructors
| Constructor | Summary |
|---|---|
MbppDataLoaderOptions() |
Properties
| Property | Summary |
|---|---|
DatasetSplit Split { get; set; } |
Dataset split. |
bool AutoDownload { get; set; } |
Auto-download. |
int MaxPromptLength { get; set; } |
Max prompt length in tokens. |
int MaxSolutionLength { get; set; } |
Max solution length in tokens. |
int VocabularySize { get; set; } |
Max vocab. |
int? MaxSamples { get; set; } |
Optional sample cap. |
string DataPath { get; set; } |
Root data path. |
Methods
| Method | Summary |
|---|---|
void Validate() |
Validates that all option values are within acceptable ranges. |