tfep.nn.transformers

Transformers for autoregressive normalizing flows in PyTorch.

All the layers defined in this module are invertible and implement an inverse() method (not to be comfused with the Tensor’s backward() method which backpropagate the gradients).

The forward propagation of the modules here return both the transformation of the input plus the log determinant of the Jacobian.

Modules

affine	Affine transformer for autoregressive normalizing flows.
mixed	A transformer applying different transformers to different features.
moebius	Moebius transformation for autoregressive normalizing flows.
quatprod	Quaternion product transformation for autoregressive normalizing flows.
sos	Sum-of-squares polynomial transformer for autoregressive normalizing flows.
spline	Circular spline transformer for autoregressive normalizing flows.
transformer	Base transformer classes for autoregressive flows.