Accelerating TensorFlow Performance on Mac
Posted by Pankaj Kanwar and Fred Alcober
With TensorFlow 2, best-in-class training performance on a variety of different platforms, devices and hardware enables developers, engineers, and researchers to work on their preferred platform. TensorFlow users on Intel Macs or Macs powered by Apple’s new M1 chip can now take advantage of accelerated training using Apple’s Mac-optimized version of TensorFlow 2.4 and the new ML Compute framework. These improvements, combined with the ability of Apple developers being able to execute TensorFlow on iOS through TensorFlow Lite, continue to showcase TensorFlow’s breadth and depth in supporting high-performance ML execution on Apple hardware.
Performance on the Mac with ML Compute
The Mac has long been a popular platform for developers, engineers, and researchers. With Apple’s announcement last week, featuring an updated lineup of Macs that contain the new M1 chip, Apple’s Mac-optimized version of TensorFlow 2.4 leverages the full power of the Mac with a huge jump in performance.
ML Compute, Apple’s new framework that powers training for TensorFlow models right on the Mac, now lets you take advantage of accelerated CPU and GPU training on both M1- and Intel-powered Macs.
For example, the M1 chip contains a powerful new 8-Core CPU and up to 8-core GPU that are optimized for ML training tasks right on the Mac. In the graphs below, you can see how Mac-optimized TensorFlow 2.4 can deliver huge performance increases on both M1- and Intel-powered Macs with popular models.
|Training impact on common models using ML Compute on M1- and Intel-powered 13-inch MacBook Pro are shown in seconds per batch, with lower numbers indicating faster training time.|
|Training impact on common models using ML Compute on the Intel-powered 2019 Mac Pro are shown in seconds per batch, with lower numbers indicating faster training time.|
Getting Started with Mac-optimized TensorFlow
Users do not need to make any changes to their existing TensorFlow scripts to use ML Compute as a backend for TensorFlow and TensorFlow Addons.
To get started, visit Apple’s GitHub repo for instructions to download and install the Mac-optimized TensorFlow 2.4 fork.
In the near future, we’ll be making updates like this even easier for users to get these performance numbers by integrating the forked version into the TensorFlow master branch.
You can learn more about the ML Compute framework on Apple’s Machine Learning website.
- Testing conducted by Apple in October and November 2020 using a preproduction 13-inch MacBook Pro system with Apple M1 chip, 16GB of RAM, and 256GB SSD, as well as a production 1.7GHz quad-core Intel Core i7-based 13-inch MacBook Pro system with Intel Iris Plus Graphics 645, 16GB of RAM, and 2TB SSD. Tested with prerelease macOS Big Sur, TensorFlow 2.3, prerelease TensorFlow 2.4, ResNet50V2 with fine-tuning, CycleGAN, Style Transfer, MobileNetV3, and DenseNet121. Performance tests are conducted using specific computer systems and reflect the approximate performance of MacBook Pro.
- Testing conducted by Apple in October and November 2020 using a production 3.2GHz 16-core Intel Xeon W-based Mac Pro system with 32GB of RAM, AMD Radeon Pro Vega II Duo graphics with 64GB of HBM2, and 256GB SSD. Tested with prerelease macOS Big Sur, TensorFlow 2.3, prerelease TensorFlow 2.4, ResNet50V2 with fine-tuning, CycleGAN, Style Transfer, MobileNetV3, and DenseNet121. Performance tests are conducted using specific computer systems and reflect the approximate performance of Mac Pro.
Related Google News:
- Scaling deep retrieval with TensorFlow Recommenders and Vertex AI Matching Engine May 1, 2023
- Unleash your Google Cloud data with ThoughtSpot, Looker, and BigQuery May 1, 2023
- Scalable electronic trading on Google Cloud: A business case with BidFX May 1, 2023
- Google Cloud and Equinix: Building Excellence in ML Operations (MLOps) May 1, 2023
- Effingo: the internal Google copy service moving data at scale May 1, 2023
- Evaluating the true cost or TCO of a database — and how Cloud Spanner compares May 1, 2023
- Google at ICLR 2023 May 1, 2023
- Serving With TF and GKE: Stable Diffusion April 28, 2023