Training times once in the 6-8 second range are now measured in fractions of a second.Īpple announces the M1, the first chip in its Apple Silicon family The change from CPU-only to CPU+GPU could account for a great deal of the improvement, as the benchmarks on an Intel-based Mac Pro show huge gains on the same hardware.

(Most large-scale ML training is done using cloud computing.) But the fact that previous versions of TF only utilized the CPU on Macs and not the powerful parallel processors in the GPU probably limited the pool of people who inflict that problem on themselves in the first place. That's worth celebrating on its own for anyone who works in ML and finds themselves constantly waiting for their models to bake.

Although a big part of that is that until now the GPU wasn't used for training tasks (!), M1-based devices see even further gains, suggesting a spate of popular workflow optimizations like this one are incoming.Īnnounced on both TensorFlow and Apple's blogs, the improved Mac version shows in the best case more than a 10x improvement in speed for common training tasks. A new Mac-optimized fork of machine learning environment TensorFlow posts some major performance increases.

Ksp mac m1