Tensorflow with proper compiler flag error message

General Discussion
#1

I have tensorflow 2.9.1 installed and using it with a GPU on Linux. when I run my code I get the following message. What is it trying to tell me?
2022-10-07 15:30:43.886417: I tensorflow/core/platform/cpu_feature_guard.cc:193] This TensorFlow binary is optimized with oneAPI Deep Neural Network Library (oneDNN) to use the following CPU instructions in performance-critical operations: AVX2 FMA
To enable them in other operations, rebuild TensorFlow with the appropriate compiler flags.

#3

This is not an error, it’s just telling you that it can and will take advantage of your CPU to get that extra speed. For more details please refer here. Thank you

1 Like
#7

@emscom,

The above is not a warning and is just a point of information.

Thank you!

#9

I hope the following question makes sense:

I would like to know if the fact that this Tensorflow is optimized with oneDNN for CPU instructions would interphere in any GPU performance, I mean, by mixing CPU and GPU instructions during a train, for instance.

Please, thanks.

2 Likes
#11

My problem is that it can take 10 seconds to create a TensorFlow tensor in PyCharm with this this check. I’m running on a desktop with GPU. The option of using conda to install a GPU version of Tensorflow seems to have been deprecated.
Thanks