mlx.nn.losses.mse_loss

class mlx.nn.losses.mse_loss(predictions: array, targets: array, reduction: str = 'none')

Computes the mean squared error loss between predictions and targets.

Parameters:

• predictions (mx.array) – The predicted values.

• targets (mx.array) – The target values.

• reduction (str, optional) – Specifies the reduction to apply to the output: 'none' | 'mean' | 'sum'. Default: 'none'.

Returns:

The computed mean squared error loss.

Return type:

mx.array