Created
December 28, 2021 00:33
-
-
Save lazyoracle/5c12bcf848243561d4556955f90fc547 to your computer and use it in GitHub Desktop.
Setting up tensorflow with GPU support on Apple Silicon
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
| # install miniforge for arm64 | |
| # https://github.com/conda-forge/miniforge | |
| name: tensorflow | |
| channels: | |
| - apple | |
| - conda-forge | |
| dependencies: | |
| - ipykernel=6.6 | |
| - jupyter=1.0 | |
| - matplotlib=3.5 | |
| - notebook=6.4 | |
| - python=3.8 | |
| - scikit-learn=1.0 | |
| - scipy=1.7 | |
| - setuptools=60.1 | |
| - tensorflow=2.6 | |
| - tensorflow-deps=2.6 | |
| - pip: | |
| - tensorflow-metal==0.3 |
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
M1 Built-in GPU support for Tensorflow is extremely rudimentary at this point and can cause serious performance degradation when compared to using Tensorflow in CPU-only mode on Apple silicon. Comment out
tensorflow-depsandtensorflow-metalin the above list to get the CPU only version.