quaterion.eval.group.group_metric module¶
- class GroupMetric(distance_metric_name: Distance = Distance.COSINE)[source]¶
Bases:
BaseMetric
Base class for group metrics
- Parameters:
distance_metric_name – name of a distance metric to calculate distance or similarity matrices. Available names could be found in
Distance
.
Provides default implementations for distance and interaction matrices calculation. Accumulates embeddings and groups in an accumulator.
- compute(embeddings: Tensor, groups: Tensor) Tensor [source]¶
Compute metric value
- Parameters:
embeddings – embeddings to calculate metrics on
groups – groups to calculate labels
- Returns:
torch.Tensor - computed metric
- static prepare_labels(groups: Tensor)[source]¶
Compute metric labels based on samples groups
- Parameters:
groups – groups to distinguish similar and dissimilar objects
- Returns:
target – torch.Tensor - labels to be used during metric computation
- raw_compute(distance_matrix: Tensor, labels: Tensor) Tensor [source]¶
Perform metric computation on ready distance_matrix and labels
This method does not make any data and labels preparation. It is assumed that distance_matrix has already been calculated, required changes such masking distance from an element to itself have already been applied and corresponding labels have been prepared.
- Parameters:
distance_matrix – distance matrix ready to metric computation
labels – labels ready to metric computation with the same shape as distance_matrix. Possible values are in {0, 1}.
- Returns:
torch.Tensor - calculated metric value