After upgrading Mac OS to Sonoma I get the following

Research & Models
#1

I had to reinstall all the TensorFlow libraries for MacOS in a new enviornment and now I am getting the following errors:
2023-09-28 10:33:11.488509: I metal_plugin/src/device/metal_device.cc:1154] Metal device set to: Apple M1 Pro
2023-09-28 10:33:11.489411: I metal_plugin/src/device/metal_device.cc:296] systemMemory: 16.00 GB
2023-09-28 10:33:11.489428: I metal_plugin/src/device/metal_device.cc:313] maxCacheSize: 5.33 GB
2023-09-28 10:33:11.489667: I tensorflow/core/common_runtime/pluggable_device/pluggable_device_factory.cc:303] Could not identify NUMA node of platform GPU ID 0, defaulting to 0. Your kernel may not have been built with NUMA support.
2023-09-28 10:33:11.489697: I tensorflow/core/common_runtime/pluggable_device/pluggable_device_factory.cc:269] Created TensorFlow device (/job:localhost/replica:0/task:0/device:GPU:0 with 0 MB memory) → physical PluggableDevice (device: 0, name: METAL, pci bus id: )
Training a brand new model - [4, 47, 53, 60]-Long
2023-09-28 10:33:17.614372: I tensorflow/core/grappler/optimizers/custom_graph_optimizer_registry.cc:114] Plugin optimizer for device_type GPU is enabled.
2023-09-28 10:33:17.622099: I tensorflow/core/grappler/optimizers/custom_graph_optimizer_registry.cc:114] Plugin optimizer for device_type GPU is enabled.
Epoch 1/10
2023-09-28 10:33:18.796718: I tensorflow/core/grappler/optimizers/custom_graph_optimizer_registry.cc:114] Plugin optimizer for device_type GPU is enabled.

Also the training (fit) is extremely slow

#3

Hi @Erez_Katz

These are only the informational messages by TensorFlow as all the logs have prefix (I). These messages will not interrupt the code execution.

You use below code on top of TensorFlow import to suppress these warnings.

import os 
os.environ['TF_CPP_MIN_LOG_LEVEL'] = '2'
import tensorflow as tf

Please share the minimal reproducible code to check the model training slowness.

#4

Thank you again Renu – I will be adding these to suppress warnings
Kind regards,

Erez