Following the instructions at
https://developer.apple.com/metal/tensorflow-plugin/
I got as far as
python -m pip install tensorflow-macos
and it responded "ERROR: Could not find a version that satisfies the requirement tensorflow-macos (from versions: none) ERROR: No matching distribution found for tensorflow-macos"
I'd be grateful for any suggestions
Thanks for your answer, but my Python is 3.8.5. Calling 'python3' causes the same error.
python3 -m pip install tensorflow-macos
ERROR: Could not find a version that satisfies the requirement tensorflow-macos (from versions: none)
ERROR: No matching distribution found for tensorflow-macos
python3 --version
Python 3.8.5
Using Pycharm CE Apple Silicon I just added in the venv with python 3.8 the packages tensorflow-macos 2.9 and tensorflow-metal 0.5.0. This runs perfectly with GPU support on my M1 Pro.
Using the latest versions tensorflow 2.11 and tensorflow-metal 0.7.0 results in error 'NOT_FOUND: could not find registered platform with id: 0x28edf1f90'. Explanation for this see here: https://developer.apple.com/forums/thread/721619
Had same problem on MacBook M1 Max. I downgraded to tensorflow-metal 0.6 and tensor flow-macos 2.10 works all well now.
brew install python@3.8
/opt/homebrew/Cellar/python@3.8/3.8.16/bin/python3.8 -m venv ~/TF
source ~/TF/bin/activate
pip install --upgrade pip
pip install tensorflow-macos
I'm following the instruction here: https://developer.apple.com/metal/tensorflow-plugin/ and it works
I get the following error after following the instructions from App Metal install website: Could not identify NUMA node of platform GPU ID 0, defaulting to 0. Your kernel may not have been built with NUMA support.
-
—
UrAvgProdigyDev
-
—
rodmclaughlin
Add a CommentI am encountering the same problem with every approach I take. Why can't you all just work in the beta and try to give us an ETA? All you have done is made the repository questions read only. This might just make me switch over to a Windows machine...
See also https://stackoverflow.com/questions/56604348/how-to-install-tensorflow-2-0-on-mac-or-linux.
This covers the efforts of several developers to use several methods to get ML working on a Mac - all have run into dead-ends.