search

sparks-baird / palette-dft-relaxation-surrogate

Pallete image processing models (e.g. JPEG restoration, inpainting), adapted to act as a surrogate model for density functional theory (DFT) relaxation of crystal structures and structure prediction.
MIT License
0 stars 0 forks source link

set of indices for composition-specific features and set of indices for structure-specific features #4

Open sgbaird opened 2 years ago

sgbaird commented 2 years ago

Sets of indices for the composition- and structure-specific features. For example, to perform structure prediction via inpainting, we'd mask the structure-specific features. Likewise, for JPEG restoration or similar to perform relaxation, it's good to be aware of which ones are related to composition to keep these fixed (e.g. via post-processing, but preferably in the JPEG restoration itself).

@michaeldalverson, curious if you could help out here. Do you mind sharing what these indices are?

@hasan-sayeed, related to this - it might be good to have functions that split/join @michaeldalverson's representation based on composition and structure. E.g. c = generate_comp_features(C) where C is a Composition object, s = generate_struct_features(S), and out = combine_comp_struct_features(c, s)

sgbaird commented 2 years ago

@hasan-sayeed if you already know how to separate these, that works too