deepmd.pt.loss.denoise

deepmd.pt.loss.denoise#

Classes#

DenoiseLoss

Helper class that provides a standard way to create an ABC using

Module Contents#

class deepmd.pt.loss.denoise.DenoiseLoss(ntypes, masked_token_loss=1.0, masked_coord_loss=1.0, norm_loss=0.01, use_l1=True, beta=1.0, mask_loss_coord=True, mask_loss_token=True, **kwargs)[source]#

Bases: deepmd.pt.loss.loss.TaskLoss

Helper class that provides a standard way to create an ABC using inheritance.

ntypes[source]#

masked_token_loss = 1.0[source]#

masked_coord_loss = 1.0[source]#

norm_loss = 0.01[source]#

has_coord[source]#

has_token[source]#

has_norm[source]#

use_l1 = True[source]#

beta = 1.0[source]#

frac_beta = 1.0[source]#

mask_loss_coord = True[source]#

mask_loss_token = True[source]#

forward(model_pred, label, natoms, learning_rate, mae=False)[source]#

Return loss on coord and type denoise.

Returns:

loss: Loss to minimize.