Community
cancel
Showing results for 
Search instead for 
Did you mean: 
Highlighted
Beginner
7 Views

oneDNN Library and python code

Hello,

I had a question on using the oneDNN framework. I am going through the developers guide, and I m having some difficulty figuring out how I can link my python code for NNs and the oneDNN library. Is there some sample code/examples I can reference for this? 

Further, how does Intel's distribution of Python come in the picture. Is it embedded with oneDNN and will any of my Intel python code automatically use oneDNN for optimization?

Just wondering how to use oneDNN to analyze and optimize my python code.

Thank you!

Tags (1)
0 Kudos
5 Replies
Highlighted
Moderator
7 Views

Hi,

Hi,

Thanks for your interest in oneDNN.

oneDNN provides a SYCL* extensions API for CPU and GPU and is available as a part of Intel® oneAPI DL Framework Developer Toolkit (https://software.intel.com/content/www/us/en/develop/tools/oneapi/dl-framework-developer-toolkit.htm...) .

You can refer sample tutorials avaialble for oneDNN at :

https://software.intel.com/content/www/us/en/develop/tools/oneapi/components/onednn.html

https://github.com/intel/BaseKit-code-samples/tree/master/oneDNN/oneDNN_Getting_Started

Regarding python code integration, we will check with our experts and get back to you.

 

 

 

0 Kudos
Highlighted
Moderator
7 Views

Hi,

Hi,

oneDNN is intended for deep learning applications and framework developers interested in improving application performance on Intel CPUs and GPUs .
Intel optimized DL frameworks like Intel Optimized Tensorflow and Pytorch is enabled with oneDNN by default.
Hence no additional intergration with your python code is required.

Thanks

 

0 Kudos
Highlighted
Moderator
7 Views

Hi,

Hi,

Could you please confirm whether the solution provided was helpful?

 

Thanks

0 Kudos
Highlighted
Beginner
7 Views

Hello,

Hello,

The solution provided was very helpful! Thank you so much!

 

- Rushiv

0 Kudos
Highlighted
Moderator
7 Views

Hi,

Hi,

Thanks for the confirmation. We are closing this case. Please feel free to open a new thread if you have any further queries.

 

0 Kudos