pub fn mae_loss<S: Shape, E: Dtype, D: Device<E>, T: Tape<E, D>>( pred: Tensor<S, E, D, T>, targ: Tensor<S, E, D> ) -> Tensor<Rank0, E, D, T>
Mean absolute error. This computes (pred - targ).abs().mean()
(pred - targ).abs().mean()
See MeanTo, abs(), and sub()