This repository was archived by the owner on May 8, 2021. It is now read-only.

Description
I know this is a small commit and i should be able to write the PR myself, however i wanted the devs input beforehand.
Currently, the GPU name is generated by appending a ".index" to every gpu. When only running one GPU i think it would be cleaner just to have the gpu name as is.
Would this be acceptable, or break the design philosophy?