So, in lecture Kayvon mentioned that there are a handful of people around the world who will be writing their own scheduling algorithms to use instead of Halide. Are there certain tasks where Halide falls short? Are these addressable challenges to actually solve, or is this a particularly intractable problem space?
So, in lecture Kayvon mentioned that there are a handful of people around the world who will be writing their own scheduling algorithms to use instead of Halide. Are there certain tasks where Halide falls short? Are these addressable challenges to actually solve, or is this a particularly intractable problem space?