...menustart

GPU Accelerate on MacOSX

...menuend

GPU Accelerate on MacOSX

PyTorch AMD (not support MacOSX)
- The official AMD instructions on building Pytorch is here
OpenCL
- GPU-Accelerated Machine Learning on MacOSX
- Deep Learning on a Mac with AMD GPU

Mac-optimized TensorFlow

CATUION: you must import mlcompute , see details below
To get started, visit Apple’s GitHub repo for instructions to download and install the Mac-optimized TensorFlow 2.4 fork.

python3.8 -m venv tf_osx
source tf_osx/bin/activate

/bin/bash -c "$(curl -fsSL https://raw.githubusercontent.com/apple/tensorflow_macos/master/scripts/download_and_install.sh)"

select compute device

# Import mlcompute module to use the optional set_mlc_device API for device selection with ML Compute.
from tensorflow.python.compiler.mlcompute import mlcompute

# Select CPU device.
mlcompute.set_mlc_device(device_name='cpu') # Available options are 'cpu', 'gpu', and ‘any'.

TensofFlow docker

docker pull tensorflow/tensorflow:latest
docker run --rm -v `pwd`:/scripts  tensorflow/tensorflow:latest python /scripts/tf_test.py