jaxopt.loss.binary_sparsemax_loss
- jaxopt.loss.binary_sparsemax_loss(label, logit)[source]
Binary sparsemax loss.
- Parameters
label (
int) – ground-truth integer label (0 or 1).logit (
float) – score produced by the model (float).
- Return type
float- Returns
loss value
References
Learning with Fenchel-Young Losses. Mathieu Blondel, André F. T. Martins, Vlad Niculae. JMLR 2020. (Sec. 4.4)