From a systems POV, the news in deep learning is that it's the first family of the long-awaited "non-word-level" APIs seeing actual real life uses. Graphics APIs devolved from OpenGL into word-level CUDA/OpenCL. Will TensorFlow do so, too, or will it make compilers evolve (MLIR)?
0 replies
1 retweet
1 like
Loading seems to be taking a while.
Twitter may be over capacity or experiencing a momentary hiccup. Try again or visit Twitter Status for more information.