vraven · function
explain_dataset
explain_dataset(model: 'Any', data: 'Any', *, targets: 'int | Sequence[int] | torch.Tensor | np.ndarray | None' = None, mode: 'str' = 'fast', class_names: 'Mapping[int, str] | Sequence[str] | None' = None, max_samples: 'int | None' = 256, fingerprint_dimensions: 'int' = 128, input_getter: 'Callable[[Any], Any] | None' = None, target_getter: 'Callable[[Any], Any] | None' = None, label_getter: 'Callable[[Any], Any] | None' = None, id_getter: 'Callable[[Any], str] | None' = None, metadata: 'dict[str, Any] | None' = None, **config_overrides: 'Any') -> 'VRavenDatasetReport'Explain a bounded set of samples and compare their reduced internal traces.
Samples are explained independently so the returned fingerprint represents one
sample rather than an aggregate over a mixed batch. This is intentionally more
expensive than a single batch trace and is bounded by ``max_samples``.
Usage
from vraven import explain_dataset
result = explain_dataset(... )Source of truthThis page is generated from the checked-out public symbol, signature, and docstring. Pin the package version for reproducible work.