Skip to content

_l1_loss

l1_loss

Calculate Mean Absolute Error between two tensors.

This method can be used with TensorFlow tensors:

true = tf.constant([[0,1,0,0], [0,0,0,1], [0,0,1,0], [1,0,0,0]])
pred = tf.constant([[0.1,0.9,0.05,0.05], [0.1,0.2,0.0,0.7], [0.0,0.15,0.8,0.05], [1.0,0.0,0.0,0.0]])
L1 = fe.backend.l1_loss(y_pred=pred, y_true=true)                                         #[0.0750, 0.1500, 0.1000, 0.0000]

true = tf.constant([[1], [3], [2], [0]])
pred = tf.constant([[2.0], [0.0], [2.0], [1.0]])
L1 = fe.backend.l1_loss(y_pred=pred, y_true=true)                                         #[1., 3., 0., 1.]

This method can be used with PyTorch tensors:

true = torch.tensor([[0,1,0,0], [0,0,0,1], [0,0,1,0], [1,0,0,0]])
pred = torch.tensor([[0.1,0.9,0.05,0.05], [0.1,0.2,0.0,0.7], [0.0,0.15,0.8,0.05], [1.0,0.0,0.0,0.0]])
L1 = fe.backend.l1_loss(y_pred=pred, y_true=true)                                         #[0.0750, 0.1500, 0.1000, 0.0000]

true = torch.tensor([[1], [3], [2], [0]])
pred = torch.tensor([[2.0], [0.0], [2.0], [1.0]])
L1 = fe.backend.l1_loss(y_pred=pred, y_true=true)                                         #[1., 3., 0., 1.]

Parameters:

Name Type Description Default
y_true Tensor

Ground truth class labels with a shape like (batch) or (batch, n_classes). dtype: int, float16, float32.

required
y_pred Tensor

Prediction score for each class, with a shape like y_true. dtype: float32 or float16.

required

Returns:

Type Description
Tensor

The L1 loss between y_true and y_pred.

Raises:

Type Description
ValueError

If y_pred is an unacceptable data type.

Source code in fastestimator/fastestimator/backend/_l1_loss.py
def l1_loss(y_true: Tensor, y_pred: Tensor) -> Tensor:
    """Calculate Mean Absolute Error between two tensors.

    This method can be used with TensorFlow tensors:
    ```python

    true = tf.constant([[0,1,0,0], [0,0,0,1], [0,0,1,0], [1,0,0,0]])
    pred = tf.constant([[0.1,0.9,0.05,0.05], [0.1,0.2,0.0,0.7], [0.0,0.15,0.8,0.05], [1.0,0.0,0.0,0.0]])
    L1 = fe.backend.l1_loss(y_pred=pred, y_true=true)                                         #[0.0750, 0.1500, 0.1000, 0.0000]

    true = tf.constant([[1], [3], [2], [0]])
    pred = tf.constant([[2.0], [0.0], [2.0], [1.0]])
    L1 = fe.backend.l1_loss(y_pred=pred, y_true=true)                                         #[1., 3., 0., 1.]
    ```

    This method can be used with PyTorch tensors:
    ```python

    true = torch.tensor([[0,1,0,0], [0,0,0,1], [0,0,1,0], [1,0,0,0]])
    pred = torch.tensor([[0.1,0.9,0.05,0.05], [0.1,0.2,0.0,0.7], [0.0,0.15,0.8,0.05], [1.0,0.0,0.0,0.0]])
    L1 = fe.backend.l1_loss(y_pred=pred, y_true=true)                                         #[0.0750, 0.1500, 0.1000, 0.0000]

    true = torch.tensor([[1], [3], [2], [0]])
    pred = torch.tensor([[2.0], [0.0], [2.0], [1.0]])
    L1 = fe.backend.l1_loss(y_pred=pred, y_true=true)                                         #[1., 3., 0., 1.]
    ```

    Args:
        y_true: Ground truth class labels with a shape like (batch) or (batch, n_classes). dtype: int, float16, float32.
        y_pred: Prediction score for each class, with a shape like y_true. dtype: float32 or float16.

    Returns:
        The L1 loss between `y_true` and `y_pred`.

    Raises:
        ValueError: If `y_pred` is an unacceptable data type.
    """
    if tf.is_tensor(y_pred):
        if tf.rank(y_pred) == 1:
            y_true = tf.expand_dims(y_true, axis=-1)
            y_pred = tf.expand_dims(y_pred, axis=-1)
        mae = tf.losses.MAE(y_true, y_pred)
    elif isinstance(y_pred, torch.Tensor):
        mae = reduce_mean(torch.nn.L1Loss(reduction="none")(y_pred, y_true), axis=-1)
    else:
        raise ValueError("Unrecognized tensor type {}".format(type(y_pred)))
    return mae