Retrieval Fall-Out

Module Interface

class torchmetrics.RetrievalFallOut(empty_target_action='pos', ignore_index=None, k=None, compute_on_step=None, **kwargs)

Computes Fall-out.

Works with binary target data. Accepts float predictions from a model output.

Forward accepts:

• preds (float tensor): (N, ...)

• target (long or bool tensor): (N, ...)

• indexes (long tensor): (N, ...)

indexes, preds and target must have the same dimension. indexes indicate to which query a prediction belongs. Predictions will be first grouped by indexes and then Fall-out will be computed as the mean of the Fall-out over each query.

Parameters

• empty_target_action (str) –

  Specify what to do with queries that do not have at least a negative target. Choose from:

  • 'neg': those queries count as 0.0 (default)

  • 'pos': those queries count as 1.0

  • 'skip': skip those queries; if all queries are skipped, 0.0 is returned

  • 'error': raise a ValueError

• ignore_index (Optional[int]) – Ignore predictions where the target is equal to this number.

• k (Optional[int]) – consider only the top k elements for each query (default: None, which considers them all)

• compute_on_step (Optional[bool]) –

  Forward only calls update() and returns None if this is set to False.

  Deprecated since version v0.8: Argument has no use anymore and will be removed v0.9.

• kwargs (Dict[str, Any]) – Additional keyword arguments, see Advanced metric settings for more info.

Raises

• ValueError – If empty_target_action is not one of error, skip, neg or pos.

• ValueError – If ignore_index is not None or an integer.

• ValueError – If k parameter is not None or an integer larger than 0.

Example

>>> from torchmetrics import RetrievalFallOut
>>> indexes = tensor([0, 0, 0, 1, 1, 1, 1])
>>> preds = tensor([0.2, 0.3, 0.5, 0.1, 0.3, 0.5, 0.2])
>>> target = tensor([False, False, True, False, True, False, True])
>>> fo = RetrievalFallOut(k=2)
>>> fo(preds, target, indexes=indexes)
tensor(0.5000)

Initializes internal Module state, shared by both nn.Module and ScriptModule.

compute()

First concat state indexes, preds and target since they were stored as lists.

After that, compute list of groups that will help in keeping together predictions about the same query. Finally, for each group compute the _metric if the number of negative targets is at least 1, otherwise behave as specified by self.empty_target_action.

Return type

Tensor

Functional Interface

torchmetrics.functional.retrieval_fall_out(preds, target, k=None)

Computes the Fall-out (for information retrieval), as explained in IR Fall-out Fall-out is the fraction of non-relevant documents retrieved among all the non-relevant documents.

preds and target should be of the same shape and live on the same device. If no target is True, 0 is returned. target must be either bool or integers and preds must be float, otherwise an error is raised. If you want to measure Fall-out@K, k must be a positive integer.

Parameters

• preds (Tensor) – estimated probabilities of each document to be relevant.

• target (Tensor) – ground truth about each document being relevant or not.

• k (Optional[int]) – consider only the top k elements (default: None, which considers them all)

Return type

Tensor

Returns

a single-value tensor with the fall-out (at k) of the predictions preds w.r.t. the labels target.

Raises

ValueError – If k parameter is not None or an integer larger than 0

Example

>>> from  torchmetrics.functional import retrieval_fall_out
>>> preds = tensor([0.2, 0.3, 0.5])
>>> target = tensor([True, False, True])
>>> retrieval_fall_out(preds, target, k=2)
tensor(1.)