Laurae (version 0.0.0.9001)

loss_MCE_grad: Mean Cubic Error (gradient function)

Description

This function computes the Mean Cubic Error loss (MCE) gradient per value provided preds and labels.

Usage

loss_MCE_grad(y_pred, y_true)

Arguments

y_pred

The predictions.

y_true

The labels.

Value

The gradient of the Cubic Error per value.

Details

Supposing: \(x = preds - labels\) Loss Formula : \(abs(x^3)\) Gradient Formula : \(3 * (x * abs(x))\) Hessian Formula : \((6 * x * x) / abs(x)\)