Skip to main content
API REFERENCE/omnibias-torch

omnibias.torch.architectures

0.5.0rc1Release candidatePython >=3.10Apache-2.0

PyTorch activation derivatives, trainable operator layers, network jets, and direct differential-operator contractions.

import omnibias.torch.architectures
Module contract & mathematical conventions

Reusable PINN architectures over the shared activation derivative tower.

Includes multilayer and Fourier-feature fields, multiscale fields, hard boundary constraints, attention jets, integral kernels, and residual networks.

Shared formulas do not imply universal floating-point identity. Verify dtype, shape, derivative-order support, and the stated numerical assumptions for the path you use.

Module source ↗Source revision 9a2d73df0573Running reference examples →
attentionModule ↗ftc_netModule ↗hardbcModule ↗integral_kernelModule ↗jetkanModule ↗joint_operatorModule ↗multiscaleModule ↗pinnModule ↗piratenetModule ↗

Imports & aliases​

These names share a definition with another public module. Follow a name for its full contract, parameters, and examples.

Import from this moduleCanonical reference
AdaptiveActivationomnibias.torch.architectures.AdaptiveActivation
AdaptiveJetMLPomnibias.torch.architectures.AdaptiveJetMLP
AffineFactoromnibias.torch.architectures.AffineFactor
AffineLiftomnibias.torch.architectures.AffineLift
AttentionJetMLPomnibias.torch.architectures.AttentionJetMLP
BoundaryMaskomnibias.torch.architectures.BoundaryMask
DeepPINNHeatomnibias.torch.architectures.DeepPINNHeat
DualFTCConfigomnibias.core.ftc.DualFTCConfig
FTCNetomnibias.torch.architectures.FTCNet
FTCNetConfigomnibias.core.ftc.FTCNetConfig
FittedJointOperatorRegressoromnibias.torch.architectures.FittedJointOperatorRegressor
FourierFeatureMLPomnibias.torch.architectures.FourierFeatureMLP
HardConstraintFieldomnibias.torch.architectures.HardConstraintField
IntegralKernelConfigomnibias.core.integral_kernel.IntegralKernelConfig
JetKANomnibias.torch.architectures.JetKAN
JetKANConfigomnibias.torch.architectures.JetKANConfig
JetMLPomnibias.torch.architectures.JetMLP
JointOperatorRegressoromnibias.torch.architectures.JointOperatorRegressor
MscaleMLPomnibias.torch.architectures.MscaleMLP
OperatorMetadataomnibias.torch.architectures.OperatorMetadata
PINNHeatomnibias.torch.architectures.PINNHeat
PirateNetomnibias.torch.architectures.PirateNet
PirateNetConfigomnibias.torch.architectures.PirateNetConfig
dirichlet_intervalomnibias.torch.architectures.dirichlet_interval
dual_ftc_lossomnibias.torch.architectures.dual_ftc_loss
edge_functionsomnibias.torch.architectures.edge_functions
fit_joint_operator_regressoromnibias.torch.architectures.fit_joint_operator_regressor
ftc_blockomnibias.torch.architectures.ftc_block
homogeneous_boxomnibias.torch.architectures.homogeneous_box
init_pirate_paramsomnibias.torch.architectures.init_pirate_params
initial_valueomnibias.torch.architectures.initial_value
integral_kernel_applyomnibias.torch.architectures.integral_kernel_apply
jetkan_from_band_planomnibias.torch.architectures.jetkan_from_band_plan
make_sirenomnibias.torch.architectures.make_siren
pirate_applyomnibias.torch.architectures.pirate_apply
pirate_featuresomnibias.torch.architectures.pirate_features
← Package overviewRunnable learning examples →