Skip to content
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

Adding Mosaic logger + logging data validated event #670

Merged
merged 10 commits into from
Oct 20, 2023

Conversation

jjanezhang
Copy link
Contributor

@jjanezhang jjanezhang commented Oct 12, 2023

Adding Mosaic Logger + data validated log #670

  1. Added mosaic logger to log foundry information for each run.
  2. Logged the timestamp for data validation complete to mosaic logger

Testing

  1. Started run from MCLI using this YAML. Checked run metadata after run finished and confirmed that data_validated timestamp is logged in the run metadata as well as all other composer logged metrics.
image

llmfoundry/callbacks/__init__.py Outdated Show resolved Hide resolved
scripts/train/train.py Outdated Show resolved Hide resolved
scripts/train/train.py Outdated Show resolved Hide resolved
@jjanezhang jjanezhang self-assigned this Oct 17, 2023
Copy link
Collaborator

@dakinggg dakinggg left a comment

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Could you please update the PR description based on the current implementation?

scripts/train/train.py Outdated Show resolved Hide resolved
scripts/train/train.py Show resolved Hide resolved
scripts/train/train.py Outdated Show resolved Hide resolved
scripts/train/train.py Outdated Show resolved Hide resolved
scripts/train/train.py Outdated Show resolved Hide resolved
@jjanezhang jjanezhang changed the title Adding Mosaic Logger + Run Events Callback Adding Mosaic Logger + data validated log Oct 18, 2023
@jjanezhang jjanezhang changed the title Adding Mosaic Logger + data validated log Adding Mosaic logger + logging data validated event Oct 18, 2023
Copy link
Contributor

@irenedea irenedea left a comment

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

LGTM! 🚀 🔥

scripts/train/train.py Outdated Show resolved Hide resolved
Copy link
Collaborator

@dakinggg dakinggg left a comment

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Before merging, can you please paste into the PR description evidence of a manual test that successfully logged the metric you are interested in?

scripts/train/train.py Outdated Show resolved Hide resolved
fixed typing for mosaicmllogger

fixed formatting, moved logging logic out of logger builder

removed callback moved all to train.py

Convert to DataSpec and add token counts that include padding (#676)

Add support for automatically registering models to UC at the end of training (#618)

removed callback from init

removed callback from init

removed generate callback

sorted imports

formatting changes
@jjanezhang jjanezhang merged commit 459947c into main Oct 20, 2023
12 checks passed
@dakinggg dakinggg deleted the jane/training-events branch November 17, 2023 06:08
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
None yet
Projects
None yet
Development

Successfully merging this pull request may close these issues.

3 participants