-
Notifications
You must be signed in to change notification settings - Fork 147
feat: list all registered schedulers (#1009) #1050
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
Open
clumsy
wants to merge
2
commits into
meta-pytorch:main
Choose a base branch
from
clumsy:feat/extend_schedulers_list
base: main
Could not load branches
Branch not found: {{ refName }}
Loading
Could not load tags
Nothing to show
Loading
Are you sure you want to change the base?
Some commits from the old base branch may be removed from the timeline,
and old review comments may become outdated.
+84
−11
Open
Changes from all commits
Commits
Show all changes
2 commits
Select commit
Hold shift + click to select a range
File filter
Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
There are no files selected for viewing
This file contains hidden or 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
This file contains hidden or 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
Oops, something went wrong.
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Outdated suggestions cannot be applied.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.
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
I took a closer look at this, and realized that we dont' need this change to get what you want.
Suppose what you want is NeMO's
[torchx.schedulers]as well as other ones that you use. Python entrypoint groups are compounding (but the keys in the groups are not). So as long as you don't have a conflict in the scheduler names, you can define your own entrypoint asand NeMO defines their set as:
the schedulers torchx ends up loading would be:
However in the absence of your entrypoint
[torchx.schedulers]torchx would only load the ones defined in NeMO (nemo_sched_1andnemo_sched_2).There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
I'm aware of this effect @kiukchung but it means there must be some package re-registering the (ideally all) built-in schedulers and keep up with the list (I remember you mentioned there's no plan to extend the list though - I was not aware when this ticket was cut).
E.g. inside a NeMo container where only NeMo-Run and TorchX dependency are installed we only get NeMo-Run schedulers (like you mentioned), i.e. no
local_cwd. As as user I would like to get NeMo-ones, built-ins and maybe add my own (just like components), but not at the cost re-registering if that makes sense.There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
are you running on a pre-built/published NeMo docker image (e.g.
nvcr.io/nvidia/nemo) directly? you don't have your workspace/project to install?There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
We don't include torchx in runtime dependencies to have a nice separation in dependency closures, so no @kiukchung
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
the fact that you want to use
local_cwdmeans that you have a direct dep ontorchxno? I'm having a hard time understanding the exact use-case to be able to offer you solutions/help. Perhaps a quick catch-up over slack or VC would help?I realize this is a long standing ask so want to help unblock you.
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
I want to use the vanilla one that is already there :) So in order for me to be able to use
local_cwdfrom within the NeMo container is either install a custom package that re-register built-ins (I don't know such package nor does it look tempting to create one), or comment out entrypoints for NeMo-Run @kiukchung