Notes on how to use the internal language of toposes in algebraic geometry

Full spectrum sheaf neural network over arbitrary CW complexes.

Sheaves library for graph abstractions based on NetworkX

My master's thesis in the realm of Graph Neural Networks. In this work we propose a novel way to precompute the sheaf Laplacian motivated by the connection that exists between the global section of a sheaf and the minimization of the Dirichlet energy via graph convolution.