Skip to content

Conversation

@lelandfy
Copy link

@lelandfy lelandfy commented Feb 2, 2023

Description

The deepwalk example currently only supports ogbl dataset. Actually other algorithms like line can run ogbn dataset. So to compare performance of different algorithms against ogb dataset, the deepwalk example should support ogbn dataset. See #5238.

Checklist

Please feel free to remove inapplicable items for your PR.

  • The PR title starts with [$CATEGORY] (such as [NN], [Model], [Doc], [Feature]])
  • I've leverage the tools to beautify the python and c++ code.
  • The PR is complete and small, read the Google eng practice (CL equals to PR) to understand more about small PR. In DGL, we consider PRs with less than 200 lines of core code change are small (example, test and documentation could be exempted).
  • All changes have test coverage
  • Code is well-documented
  • To the best of my knowledge, examples are either not affected by this change, or have been fixed to be compatible with this change
  • Related issue is referred in this PR
  • If the PR is for a new model/paper, I've updated the example index here.

Changes

  • Add ogbn dataset following the style of ogbl dataset which is already supported in this example.

@dgl-bot
Copy link
Collaborator

dgl-bot commented Feb 2, 2023

Not authorized to trigger CI. Please ask core developer to help trigger via issuing comment:

  • @dgl-bot

@dgl-bot
Copy link
Collaborator

dgl-bot commented Feb 2, 2023

Commit ID: b7d3f56

Build ID: 1

Status: ❌ CI test failed in Stage [Authentication].

Report path: link

Full logs path: link

@rudongyu rudongyu self-requested a review February 3, 2023 10:38
Copy link
Collaborator

@rudongyu rudongyu left a comment

Choose a reason for hiding this comment

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

Thanks for your contribution! I've added some comments.

Besides, we also need to update the evaluation guide and the results related to ogbn data in readme. Do you have resources to do that?

fast_neg=args.fast_neg,
ogbl_name=args.ogbl_name,
load_from_ogbl=args.load_from_ogbl,
ogbn_name=args.ogbn_name,
Copy link
Collaborator

Choose a reason for hiding this comment

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

It seems the logic of loading data from ogbn is similar to that from ogbl. How about using load_from_ogb and ogb_name to unify the entrance and branching in load_from_ogb_with_name function according to the prefix of dataset name?

"ogbl-collab",
"ogbl-ddi",
"ogbl-ppa",
"ogbl-citation",
Copy link
Collaborator

Choose a reason for hiding this comment

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

Suggested change
"ogbl-citation",
"ogbl-citation2",

elif load_from_ogbn:
assert (
len(gpus) == 1
), "ogb.linkproppred is not compatible with multi-gpu training."
Copy link
Collaborator

Choose a reason for hiding this comment

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

ogb.nodeproppred?

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