Can you enable Intel Xeon Gold 6138 processor to run AVX instructions? Attempting to run the Google Python TensorFlow package of 1.9 -doesn't work tensorFlow 1.5 does (doesn't require AVX instructions)

[Emeth_O_Intel]

Hello JLaut2,

Thank you for contacting Intel Xeon Community Support.

I was reviewing your thread and I would like to gather more information in order to have a better understanding of this scenario.

• Have you noticed an “illegal instruction” error or a specific error while trying to run the TensorFlow 1.9?
• Do you know if there is a specific instruction from the AVX set that the new version is requiring?

For more information about the TensorFlow* Installation Guide, please check the following link:

• https://software.intel.com/en-us/articles/intel-optimization-for-tensorflow-installation-guide#

Please let me know if the information helps you to have, or if you still have more questions do not hesitate and let me know and I will be more than happy to assist you.

Best regards,

Emeth O.

Intel Server Specialist.

[JLaut2]

thank you for your reply, the easiest way to duplicate my issue is to run this test script:

https://gist.github.com/mrry/ee5dbcfdd045fa48a27d56664411d41c

some additional information is this 'stack overflow':

https://stackoverflow.com/questions/43751418/tensorflow-on-windows-importerror-no-module-named-pywrap-tensorflow-internal

Any help you could give me would be very much appreciated

[JLaut2]

I am using Python interpreter 3.6.5

[Emeth_O_Intel]

Hello JLaut2,

Thank you for replying back with those details, let me verify the information in order to proceed with the next step.

Best regards,

Emeth O.

Intel Server Specialist.

[Emeth_O_Intel]

Hello JLaut2,

I was verifying some details and the best way to get the support it would be at this link:

https://software.intel.com/en-us/forums/intel-isa-extensions.

You can ask those questions in that community and they will be more than happy to assist you.

Have a wonderful day.

Best regards,

Emeth O.

Intel Server Specialist.

[JLaut2]

Okay, thank you for all your efforts