tf.keras.losses.tversky

Computes the Tversky loss value between y_true and y_pred.

tf.keras.losses.tversky(
    y_true, y_pred, alpha=0.5, beta=0.5, axis=None
)

This loss function is weighted by the alpha and beta coefficients that penalize false positives and false negatives.

With alpha=0.5 and beta=0.5, the loss value becomes equivalent to Dice Loss.

Args
y_true	tensor of true targets.
y_pred	tensor of predicted targets.
alpha	coefficient controlling incidence of false positives.
beta	coefficient controlling incidence of false negatives.
axis	tuple for which dimensions the loss is calculated.

Returns
Tversky loss value.

Reference:

• Salehi et al., 2017