HumanEvalDataLoader

Models & Types · AiDotNet.Reasoning.Benchmarks.Data

Loader for HumanEval benchmark dataset.

How It Works

HumanEval format: { “task_id”: “HumanEval/0”, “prompt”: “def has_close_elements(numbers, threshold):\n """ Check if in given list…”, “canonical_solution”: “ for idx, elem in enumerate(numbers):…“, “test”: “def check(candidate):…”, “entry_point”: “has_close_elements” }

Constructors

Constructor Summary
HumanEvalDataLoader()

Methods

Method Summary
List<HumanEvalProblem> GetSampleProblems()
Task<List<HumanEvalProblem>> LoadFromFileAsync(string filePath)