TF 2.16.1 and Apple Silicon GPU support not working. about tensorflow HOT 4 OPEN

Hi @msusol ,

As a temporary fix you can use Keras2 backend. Since TF2.16 defaults to Keras3 backend it uses Keras3 by default with tf.keras. You can switch to Keras2 backend by installing tf_keras package and setting os.environ['TF_USE_LEGACY_KERAS'] = '1'. Meanwhile team will look into the issue with Keras3 vs Mac compatibility.

from tensorflow.

You will have to install tensorflow-metal to add support for GPU on Apple silicons. More info

from tensorflow.

I did. Please look at the notebooks where I show exactly what was installed.

https://github.com/msusol/jupyter-notebook-on-macos/blob/main/notebooks/TensorFlow216.ipynb

from tensorflow.

Hi @msusol ,

As a temporary fix you can use Keras2 backend. Since TF2.16 defaults to Keras3 backend it uses Keras3 by default with tf.keras. You can switch to Keras2 backend by installing tf_keras package and setting os.environ['TF_USE_LEGACY_KERAS'] = '1'. Meanwhile team will look into the issue with Keras3 vs Mac compatibility.

Verified that this change does restore GPU access to the kernel:
msusol/jupyter-notebook-on-macos@7688a9f

Waiting for the resolution to remove the legacy flag.

from tensorflow.