Open shubhang12 opened 4 years ago
On a gpu deployment, the major throttle seems to be only one core support for preprocessing of the files, I for example am running a 12 core system and could benefit a whole lot from using the unused cores
On a gpu deployment, the major throttle seems to be only one core support for preprocessing of the files, I for example am running a 12 core system and could benefit a whole lot from using the unused cores