Bachelor's Thesis: Implementing DISTANA in PyTorch Geometric and Deep Graph Library
Posted Updated 
Results of the Traffic Light Detection model on a batch of images.
By Magnus Kaut 1 min read
Thesis: “Implementing DISTANA in PyTorch Geometric and Deep Graph Library”
The goals of this work are twofold: (a) make DISTANA available on the increasingly popular APIs and to investigate PyTorch Geometric’s and Deep Graph Library’s usability when it comes to implementing a self-developed model, and (b) compare DISTANA to other GNN models in PyTorch Geometric and Deep Graph Library in terms of accuracy and computational complexity (i.e. processing speed) on different benchmarks.
This post is licensed under CC BY 4.0 by the author.