sup3r.utilities.loss_metrics.MaterialDerivativeLoss

sup3r.utilities.loss_metrics.MaterialDerivativeLoss#

class MaterialDerivativeLoss(gen_features)[source]#

Bases: Sup3rLoss

Loss class for the material derivative. This is the left hand side of the Navier-Stokes equation and is equal to internal + external forces divided by density in general. Under certain simplifying assumptions, this is equal to zero.

References

https://en.wikipedia.org/wiki/Material_derivative

Initialize the loss with given input features

Parameters:

  • gen_features (list | str) – List of generator output features that the loss metric will be calculated on. If ‘all’, the loss will be calculated on all generator features. Otherwise, the loss will be calculated on the features specified in the list. The order of features in the list will be checked to determine the order of features in the generator output tensor.

  • true_features (list | str) – List of true features that the loss metric will be calculated on. If None, this will be the same as gen_features. The order of features in the list will be checked to determine the order of features in the ground truth tensor.

Methods

call(x_true, x_gen)

Custom content loss that encourages accuracy of the material derivative.

from_config(config)

get_config()

Attributes

LOSS_METRIC

dtype

call(x_true, x_gen)[source]#

Custom content loss that encourages accuracy of the material derivative.

Parameters:

  • x_true (tf.tensor) – high resolution ground truth data (n_observations, spatial_1, spatial_2, temporal, features)

  • x_gen (tf.tensor) – synthetic generator output (n_observations, spatial_1, spatial_2, temporal, features)

Returns:

tf.tensor – 0D tensor with loss value

__call__(y_true, y_pred, sample_weight=None)#

Call self as a function.

Contents