megengine.functional.nn.l1_loss

l1_loss(pred, label, reduction='mean')[source]

Calculates the mean absolute error (MAE) between each element in the pred \(x\) and label \(y\).

The mean absolute error can be described as:

\[\ell(x,y) = mean\left(L \right)\]

where

\[L = \{l_1,\dots,l_N\}, \quad l_n = \left| x_n - y_n \right|,\]

\(x\) and \(y\) are tensors of arbitrary shapes with a total of \(N\) elements each. \(N\) is the batch size.

Parameters

  • pred (Tensor) – predicted result from model.

  • label (Tensor) – ground truth to compare.

  • reduction (str) – the reduction to apply to the output: ‘none’ | ‘mean’ | ‘sum’.

Return type

Tensor

Returns

loss value.

Shape:

  • pred: \((N, *)\) where \(*\) means any number of additional dimensions.

  • label: \((N, *)\). Same shape as pred.

Examples

>>> pred = Tensor([3, 3, 3, 3])
>>> label = Tensor([2, 8, 6, 1])
>>> F.nn.l1_loss(pred, label)
Tensor(2.75, device=xpux:0)
>>> F.nn.l1_loss(pred, label, reduction="none")
Tensor([1 5 3 2], dtype=int32, device=xpux:0)
>>> F.nn.l1_loss(pred, label, reduction="sum")
Tensor(11, dtype=int32, device=xpux:0)

previous

megengine.functional.nn.embedding

next

megengine.functional.nn.square_loss