Enter training time for BF16 using AVX512

[jafraustro]

Update IntelPyTorch training notebook to include environment variable settings for AVX512 and BF16 optimizations

Existing Sample Changes

Description

Update IntelPyTorch training notebook to include environment variable settings for AVX512 and BF16 optimizations

How Has This Been Tested?

Please describe the tests that you ran to verify your changes. Provide instructions so we can reproduce. Please also list any relevant details for your test configuration

• Command Line
• oneapi-cli
• Visual Studio
• Eclipse IDE
• VSCode
• When compiling the compliler flag "-Wall -Wformat-security -Werror=format-security" was used

[Ankur-singh]

@jafraustro can you please verify if the environment variable is actually affecting the trainModel function?

It would to ideal to have the trainModel function in a separate script. This would allow use to do the following, making sure that the env variable is initialized properly:

# In jupyter notebook cell
os.environ["ONEDNN_MAX_CPU_ISA"] = "AVX512_CORE_BF16"
!python pytorch_training_avx512_bf16.py

I also see pytorch_training_amx_bf16.py and pytorch_training_avx512_bf16.py scripts with mostly the same code. Even these scripts can be cleaned. Let me know what you think?

[jafraustro]

Hi @Ankur-singh,

You are right about the ONEDNN_MAX_CPU_ISA variable.

I made the following changes:

• The file pytorch_training_amx_bf16.py has been removed since the code is in the notebook.
• The file pytorch_training_avx512_bf16.py is used to run the same code, but it sets the environment variable os.environ["ONEDNN_MAX_CPU_ISA"] = "AVX512_CORE_BF16".
• The variable bf16_avx512_training_time is now saved in a file for use in the notebook.