# torch_geometric.explain.metric.fidelity_curve_auc

fidelity_curve_auc(pos_fidelity: Tensor, neg_fidelity: Tensor, x: Tensor) [source]

Returns the AUC for the fidelity curve as described in the “GraphFramEx: Towards Systematic Evaluation of Explainability Methods for Graph Neural Networks” paper.

More precisely, returns the AUC of

$f(x) = \frac{\textrm{fid}_{+}}{1 - \textrm{fid}_{-}}$
Parameters
• pos_fidelity (torch.Tensor) – The positive fidelity $$\textrm{fid}_{+}$$.

• neg_fidelity (torch.Tensor) – The negative fidelity $$\textrm{fid}_{-}$$.

• x (torch.Tensor) – Tensor containing the points on the $$x$$-axis. Needs to be sorted in ascending order.