Closed Zymrael closed 1 year ago
Adds discrete cosine transforms (type-I and II for now) and an nn.Module API. Explicit transforms precompute the matrix for efficiency at a memory cost. Starting point: https://github.com/zh217/torch-dct/blob/master/torch_dct/_dct.py.
nn.Module
Explicit
This is prep before adding neural operator and T1 layers, here and in diffeqml-models.
diffeqml-models
The goal is to have a solid selection of transforms in functional and nn.Module format that can be plugged into neural operator layers.
DFTConvResidual
SpectralConv
T1
@fedebotu to follow in case you are interested in playing around with some other transforms previously discussed.
Adds discrete cosine transforms (type-I and II for now) and an
nn.Module
API.Explicit
transforms precompute the matrix for efficiency at a memory cost. Starting point: https://github.com/zh217/torch-dct/blob/master/torch_dct/_dct.py.This is prep before adding neural operator and T1 layers, here and in
diffeqml-models
.The goal is to have a solid selection of transforms in functional and
nn.Module
format that can be plugged into neural operator layers.DFTConvResidual
layers (SpectralConv
) of FNOsT1
layers@fedebotu to follow in case you are interested in playing around with some other transforms previously discussed.