The experimental data coming from the camera and the detector have a resolution that is captured by float16 format.
This means that the model does not have to be more precise than that and can be made faster by using either float16 (if compatible) or at least float32.
The experimental data coming from the camera and the detector have a resolution that is captured by
float16
format.This means that the model does not have to be more precise than that and can be made faster by using either
float16
(if compatible) or at leastfloat32
.