ApproxLossFunction

class lightning_ir.loss.approximate.ApproxLossFunction(temperature: float = 1)[source]

Bases: ListwiseLossFunction

Base class for approximate loss functions that compute ranks from scores.

__init__(temperature: float = 1) → None[source]

Initialize the ApproxLossFunction.

Parameters:

temperature (float) – Temperature parameter for scaling the scores. Defaults to 1.

Methods

__init__([temperature])

Initialize the ApproxLossFunction.

get_approx_ranks(scores, temperature)

Compute approximate ranks from scores.
static get_approx_ranks(scores: Tensor, temperature: float) → Tensor[source]

Compute approximate ranks from scores.

Parameters:

  • scores (torch.Tensor) – The input scores.

  • temperature (float) – Temperature parameter for scaling the scores.

Returns:

The computed approximate ranks.

Return type:

torch.Tensor