Apple debuts a forked version of TensorFlow optimized for macOS, says it trains up to 7x faster on 13" MacBook Pro with M1 than 2020 Macbook Pro 13" with Intel
Join GitHub today — GitHub is home to over 50 million … Mayank Sharma / TechRadar : Apple M1 Macs will make training AI much faster Ewdison Then / SlashGear : Google TensorFlow ML framework gets an Apple M1-optimized version Andrei Frumusanu / AnandTech : The 2020 Mac Mini Unleashed: Putting Apple Silicon M1 To The Test Brian Fagioli / BetaNews : Apple M1 Mac mini already discounted for holiday shopping — a great Christmas gift! Serdar Yegulalp / InfoWorld : Apple releases TensorFlow fork with speedups for M1 Macs Tweets: @tensorflow : ⚡️ Accelerating TensorFlow 2 performance on Mac @Apple's new Mac-optimized TensorFlow 2.4 fork lets you speed up training on Macs, resulting in up to 7x faster performance on platforms with the new M1 chip! Learn how ↓ https://blog.tensorflow.org/ ... Max Woolf / @minimaxir : Apple released a fork of TensorFlow that allows Macs to train models using the M1. https://machinelearning.apple.com/ ... The performance benchmarks against GPU-training are indeed impressive. https://twitter.com/... Rene Ritchie / @reneritchie : [New tensorflow_macos fork of TensorFlow 2.4 leverages ML Compute to enable ML libraries to take full advantage of not only the CPU, but also the GPU in both M1- and Intel Macs for dramatically faster training performance.] https://machinelearning.apple.com/ ... https://blog.tensorflow.org/ ... Elie Bursztein / @elie : Yay! TensorFlow is now accelerated on MacOS https://blog.tensorflow.org/ ... The gains on the new Apple M1 are particularly impressive. #TensorFlow #macOS #MachineLearning #AI https://twitter.com/... Max Woolf / @minimaxir : The first chart specifically is on a Mac Pro alone (not the M1), meaning that Apple somehow found a way to get GPU training working on a Mac with TF. https://github.com/... If Apple could help port that code to PyTorch that might change the game a bit. Sandeep Srinivasa / @sandeepssrin : Holy crap. @TensorFlow 2.4 supports “ML Compute” on Apple M1. This means using GPU for acceleration. Like Nvidia's CUDA monopoly. Apple couldn't get this for their Intel CPU ? It somehow points the blame at Intel as the villian all along. https://machinelearning.apple.com/ ... Wl Aknfadrn / @waleakinfaderin : Pretty interesting. They achieved 3x-7x faster training using the Mac-optimized TensorFlow framework on the 13-inch MacBook Pro with M1 (comparison with Intel chips). Leveraging ML Compute for Accelerated Training on Mac - Apple Machine Learning Research https://machinelearning.apple.com/ ... Stephen Shankland / @stshank : Huge performance increase of Google's TensorFlow AI software (open source and many others use it too) on Apple's new M1 Macs. (Shorter is faster on this chart.) Uses Apple's ML Compute interface for AI training that speeds up ML on Intel Macs, too. https://blog.tensorflow.org/ ... https://twitter.com/...