Every story and comment from the Hacker News API, in a layout borrowed from 2001.
A comment on Render a neural network into CUDA/HIP code. Parent comment
I brought it up because shuffling weights around takes lots of time, takes up more RAM, and saps memory bandwidth the IGP and CPU desparately need.
There are no comments on this comment yet.
I brought it up because shuffling weights around takes lots of time, takes up more RAM, and saps memory bandwidth the IGP and CPU desparately need.