dataeval.core.rerank_hard_first¶

dataeval.core.rerank_hard_first(result)¶

Reverse ranking order to put hard samples first.

Takes a RankResult (expected to be in easy_first order) and reverses the indices to produce hard_first order.

Parameters:¶

result : RankResult¶

Ranking result, typically in easy_first order.

Returns:¶

Dictionary containing:

• indices: NDArray[np.intp] - Reversed indices (hard samples first)

• scores: NDArray[np.float32] | None - Scores in original order (unchanged if present)

• method: str - Same as input

• policy: str - “hard_first”

Return type:¶

RankResult

Examples

>>> from dataeval.core import rank_knn, rerank_hard_first
>>> import numpy as np
>>> embeddings = np.random.rand(100, 64).astype(np.float32)
>>> result = rank_knn(embeddings, k=5)
>>> result = rerank_hard_first(result)