[Question] TransfomerEngine and Apex dependencies

[peri044]

Describe the bug

The Dockerfile is using 24.03 which comes with TransformerEngine and apex (via torch.cuda.apex). (Release notes). But it seems like transformers engine and apex are being installed from source.

1. I was wondering if this is required ?
2. Is it meant to override the existing versions?
3. Also, is there a reason to use 24.03 instead of more recent ones ?

cc: @odelalleau

Steps/Code to reproduce bug

Please list minimal steps or code snippet for us to be able to reproduce the bug.

A helpful guide on on how to craft a minimal bug report http://matthewrocklin.com/blog/work/2018/02/28/minimal-bug-reports.

Expected behavior

A clear and concise description of what you expected to happen.

Environment overview (please complete the following information)

• Environment location: [Bare-metal, Docker, Cloud(specify cloud provider - AWS, Azure, GCP, Collab)]
• Method of NeMo-Aligner install: [pip install or from source]. Please specify exact commands you used to install.
• If method of install is [Docker], provide docker pull & docker run commands used

Environment details

If NVIDIA docker image is used you don't need to specify these.
Otherwise, please provide:

• OS version
• PyTorch version
• Python version

Additional context

Add any other context about the problem here.
Example: GPU model