[docs] Performance docs refactor p.2 #26791
Merged
Conversation
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
|
The documentation is not available anymore as the PR was closed or merged. |
younesbelkada
approved these changes
Oct 18, 2023
stevhliu
reviewed
Oct 19, 2023
There was a problem hiding this comment.
Thanks for your work on this massive doc! Some thoughts:
- I feel like the Concepts section is a bit redundant when we go into much more detail about these techniques in their individual sections. Maybe we can add these terms to the Glossary?
- From the DP + PP section to the FlexFlow section, none of these techniques are supported or integrated in Transformers, so I'm not really sure we need to add these here. I feel like it would be frustrating for a user coming to the doc expecting to use DP + PP only to see that it's not implemented. I think putting them into a new "Parallelism" doc in the conceptual guides section may be better.
|
|
||
| 1. DP+PP+TP+ZeRO-1 | ||
| 🤗 Transformers status: not yet integrated. We already have our models FX-trace-able via [transformers.utils.fx](https://github.com/huggingface/transformers/blob/master/src/transformers/utils/fx.py), which is |
There was a problem hiding this comment.
I think it's cool to mention FlexFlow, but if it isn't integrated in Transformers then I'm not sure we need to add this section here because it doesn't really help users who want to try it out but it's not available.
Also not really a fan of "so someone needs to figure out what needs to be done to make FlexFlow work with our models". The tone sounds a bit demanding and not as friendly (totally understand this was already here before you started working on it 🙂 ).
There was a problem hiding this comment.
Rephrased the sentence. I'm on the fence about removing the content, or moving it to another section.
Co-authored-by: Steven Liu <59462357+stevhliu@users.noreply.github.com>
staghado
pushed a commit
to staghado/transformers
that referenced
this pull request
Oct 24, 2023
* initial edits * improvements for clarity and flow * improvements for clarity and flow, removed the repetead section * removed two docs that had no content * Revert "removed two docs that had no content" This reverts commit e98fa2f. * Apply suggestions from code review Co-authored-by: Steven Liu <59462357+stevhliu@users.noreply.github.com> * feedback addressed * more feedback addressed * feedback addressed --------- Co-authored-by: Steven Liu <59462357+stevhliu@users.noreply.github.com>
i4never
pushed a commit
to i4never/transformers
that referenced
this pull request
Oct 25, 2023
* initial edits * improvements for clarity and flow * improvements for clarity and flow, removed the repetead section * removed two docs that had no content * Revert "removed two docs that had no content" This reverts commit e98fa2f. * Apply suggestions from code review Co-authored-by: Steven Liu <59462357+stevhliu@users.noreply.github.com> * feedback addressed * more feedback addressed * feedback addressed --------- Co-authored-by: Steven Liu <59462357+stevhliu@users.noreply.github.com>
EduardoPach
pushed a commit
to EduardoPach/transformers
that referenced
this pull request
Nov 19, 2023
* initial edits * improvements for clarity and flow * improvements for clarity and flow, removed the repetead section * removed two docs that had no content * Revert "removed two docs that had no content" This reverts commit e98fa2f. * Apply suggestions from code review Co-authored-by: Steven Liu <59462357+stevhliu@users.noreply.github.com> * feedback addressed * more feedback addressed * feedback addressed --------- Co-authored-by: Steven Liu <59462357+stevhliu@users.noreply.github.com>
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
This PR continues performance docs refactor in the transformers docs. It focuses mainly on the "Efficient Training on Multiple GPUs" doc and contains the following changes: