ROCm / ROCm/TransformerEngine

[Out of Box Experience]: ROCm Transformer Engine Should Be Included in AMD Pytorch Images

Open
#82 14 comments 1 reaction 0 assignees View on GitHub

Nobody has claimed this yet.

Under Investigation
Dominant language
Python
Stars
76
Forks
39
Avg merge
5d 23h
Merged PRs (30d)
15

Description

Suggestion Description

Hi @hliuca ,

On Nvidia NGC Pytorch Containers nvcr.io/nvidia/pytorch:24.xx-py3, Transformer Engine is included out of the box. This leads to less end user installation misconfiguration issues such as not using the correct build flags.

Currently on rocm/pytorch& on rocm/pytorch-nightly, Transformer Engine is not included out of the box.

It would be great to have parity with Nvidia on this.

https://docs.nvidia.com/deeplearning/frameworks/pytorch-release-notes/rel-24-09.html

image

Operating System

Ubuntu

GPU

MI300X

ROCm Component

No response

Contributor guide

Open the contributing guide

First steps

  1. Read the whole issue, then the project's contributing guide.
  2. Comment on the issue to say you are picking it up — it saves two people doing the same work.
  3. Fork the repository and make your change on a branch.
  4. Open a pull request that references the issue number.

Research direction

Start by reviewing how the rocm/pytorch and rocm/pytorch-nightly images are defined and how Transformer Engine is installed for AMD GPUs. Confirm the required build flags and image packaging approach, then verify that Transformer Engine is available out of the box in both images without end-user installation.

Written by the indexing model from the issue text.

Assessment

Tech stack
python, pytorch
Domain
devops, machine-learning
Issue type
Feature
Difficulty
4/5
Estimated time
3-5 days
Activity status
Stale
Clarity
Mostly clear
Newbie friendliness
35/100

Get new issues in your inbox

A short digest of beginner-friendly GitHub issues.