Skip to content

Conversation

@ECMGit
Copy link
Contributor

@ECMGit ECMGit commented Sep 15, 2024

reopened #9591

Feature summary:

  • Add GLEM as GNN & LLM Co-training model to PyG
  • adapt GLEM's LM to AutoModelForSequenceClassification from transformers
  • Lora support
  • LM/LLM support
  • ogbn-products/ogbn-arxiv testing finished
  • TAGDataset can be used as a wrapper class for any node classification dataset in PyG with LM tokenizer and associate raw text
  • external prediction as pseudo labels supported

@codecov
Copy link

codecov bot commented Sep 15, 2024

Codecov Report

Attention: Patch coverage is 11.93182% with 155 lines in your changes missing coverage. Please review.

Project coverage is 86.91%. Comparing base (f5c8293) to head (7c648f9).

Files with missing lines Patch % Lines
torch_geometric/nn/models/glem.py 11.42% 155 Missing ⚠️
Additional details and impacted files
@@            Coverage Diff             @@
##           master    #9662      +/-   ##
==========================================
- Coverage   88.13%   86.91%   -1.23%     
==========================================
  Files         482      483       +1     
  Lines       31421    31592     +171     
==========================================
- Hits        27693    27458     -235     
- Misses       3728     4134     +406     

☔ View full report in Codecov by Sentry.
📢 Have feedback on the report? Share it here.

@puririshi98 puririshi98 self-requested a review September 16, 2024 15:27
Copy link
Contributor

@puririshi98 puririshi98 left a comment

Choose a reason for hiding this comment

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

LGTM just get CI green

@puririshi98 puririshi98 marked this pull request as ready for review September 24, 2024 19:28
@puririshi98
Copy link
Contributor

@rusty1s @akihironitta ready for your reviews

Copy link
Member

@akihironitta akihironitta left a comment

Choose a reason for hiding this comment

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

Could we have type annotations all over the PR? Also, I'd suggest splitting this PR into smaller ones.

Copy link
Member

@akihironitta akihironitta left a comment

Choose a reason for hiding this comment

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

I haven't had a look outside the example script yet, but this addition is exciting! 🚀

@puririshi98
Copy link
Contributor

LGTM @akihironitta @rusty1s let us know if anything else needed

@puririshi98
Copy link
Contributor

merging this along with community sprint PRs since matthias is busy till end of year and he told me to test/review/merge them myself

@puririshi98 puririshi98 enabled auto-merge (squash) November 19, 2024 20:11
@puririshi98 puririshi98 merged commit e1a925b into pyg-team:master Nov 19, 2024
16 checks passed
mattjhayes3 pushed a commit to mattjhayes3/pytorch_geometric that referenced this pull request Dec 14, 2024
reopened  pyg-team#9591 

Feature summary:

- Add GLEM as GNN & LLM Co-training model to PyG
- adapt GLEM's LM to AutoModelForSequenceClassification from
transformers
- Lora support
- LM/LLM support
- ogbn-products/ogbn-arxiv testing finished
- TAGDataset can be used as a wrapper class for any node classification
dataset in PyG with LM tokenizer and associate raw text
- external prediction as pseudo labels supported

---------

Co-authored-by: pre-commit-ci[bot] <66853113+pre-commit-ci[bot]@users.noreply.github.com>
Co-authored-by: Rishi Puri <[email protected]>
Co-authored-by: Akihiro Nitta <[email protected]>
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Projects

None yet

Development

Successfully merging this pull request may close these issues.

3 participants