Every story and comment from the Hacker News API, in a layout borrowed from 2001.
A comment on Speeding up PyTorch inference on Apple devices with AI-generated Metal kernels. Parent comment
Back in the day TensorFlow had tfdeploy which compiled TensorFlow terms into NumPy matrix operations. Our synthetic tests saw speedups of factor 50.
There are no comments on this comment yet.
Back in the day TensorFlow had tfdeploy which compiled TensorFlow terms into NumPy matrix operations. Our synthetic tests saw speedups of factor 50.