DISCONTINUATION OF PROJECT

This project will no longer be maintained by Intel.
Intel has ceased development and contributions including, but not limited to, maintenance, bug fixes, new releases, or updates, to this project.
Intel no longer accepts patches to this project.
If you have an ongoing need to use this project, are interested in independently developing it, or would like to maintain patches for the open source software community, please create your own fork of this project.

PyTorch_Optimizations

NumPy* and PyTorch* Optimizations

Objective:

Describe how Intel SIMD and Cache optimization provided by Intel oneMKL-DNN as well as the Intel Extensions for PyTorch can accelerate your pytorch workloads especially prior to training loop or during post processing. Also explore how to use Intel Extensions to PyTorch and how to access Intel GPU for PyTorch

Replacing Inefficient code

Preparation to run on Intel DevCloud

• mkdir NumPyPyTorchOpt
• cd NumPyPyTorchOpt
• source source /opt/intel/oneapi/setvars.sh > /dev/null 2>&1 --force
• conda activate pytorch
• git clone https://github.com/IntelSoftware/PyTorch_Optimizations.git
• cd PyTorch_Optimizations

This workshop is designed to be used on the DevCloud and includes details on submitting batch jobs on the DevCloud environment.

Survey

Please fill out our survey to let us know hwo we can imporve or how it might impact your developer skills

https://intel.az1.qualtrics.com/jfe/form/SV_9uHKSYFaiC9Xjeu

License

Code samples are licensed under the MIT license. See License.txt for details. Third party program Licenses can be found here: third-party-programs.txt