Skip to content

An implementation of Neural Implicit Flow (Pan et al. - 2022) using PyTorch.

Notifications You must be signed in to change notification settings

2ailesB/NIF_torch

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

49 Commits
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 

Repository files navigation

NIF_torch

An implementation of Neural Implicit Flow (Pan et al. - 2022) using PyTorch. Tensorflow/keras version at : https://github.com/pswpswpsw/nif.

NIF architecture

Results

simple NIF

Using the cfg file 'nif_1dwave.yaml' in the config folder, we have the following result :

Loss Visual results
Training loss with last layer nif on the cylinder dataset Training results with last layer nif on the cylinder dataset
Testing loss with last layer nif on the cylinder dataset Testing results with last layer nif on the cylinder dataset

multiscale NIF

Using the cfg file 'nifmultiscale.yaml' in the config folder, we have the following result :

Loss Visual results
Training loss with last layer nif on the cylinder dataset Training results with last layer nif on the cylinder dataset
Testing loss with last layer nif on the cylinder dataset Testing results with last layer nif on the cylinder dataset

last layer NIF

Using the cfg file 'niflastlayer_cylinder.yaml' in the config folder, we have the following result :

Loss Visual results
Training loss with last layer nif on the cylinder dataset Training results with last layer nif on the cylinder dataset
Testing loss with last layer nif on the cylinder dataset Testing results with last layer nif on the cylinder dataset

Key References

[1] Neural Implicit Flow : a mesh-agnostic dimensionality reduction of spatio-temporal data, Shaowu Pan, Steven L. Brunton, J. Nathan Kutz, 2022

About

An implementation of Neural Implicit Flow (Pan et al. - 2022) using PyTorch.

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published

Languages