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.

Sign up for GitHub

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

Jump to bottom

Support Negative Sampling in pylibcugraph and cuGraph-PyG #4660

Merged
merged 39 commits into from
Sep 30, 2024
Merged

Support Negative Sampling in pylibcugraph and cuGraph-PyG #4660

merged 39 commits into from
Sep 30, 2024

Conversation

alexbarghi-nv
Copy link
Member

@alexbarghi-nv alexbarghi-nv commented Sep 20, 2024 •
edited
Loading

Adds support for negative sampling in pylibcugraph and cugraph-pyg.

Closes rapidsai/cugraph-gnn#39
Merge after #4641

alexbarghi-nv and others added 29 commits August 21, 2024 08:23
@alexbarghi-nv
sample buffer
c386402
@alexbarghi-nv
sampling
d8916c6
@alexbarghi-nv
fix imports
1fcd982
@alexbarghi-nv
import torch
a22ce90
@alexbarghi-nv
fix bug, write test
55d8ad1
@alexbarghi-nv
mg testing
5e1b986
@alexbarghi-nv
update examples
a365a24
@alexbarghi-nv
temporarily add nvtx annotation
ffc0e46
@alexbarghi-nv
remove annotations
00e28c1
@alexbarghi-nv
merge
4571711
@alexbarghi-nv
revert change to dgl sampler
3131926
@alexbarghi-nv
revert newline
b92c30b
@alexbarghi-nv
expose buffered sample reader
62a2fdf
@alexbarghi-nv
Merge branch 'sample-buffer' into pyg-linkpred
6553464
@alexbarghi-nv
initial write
5f21d2b
@alexbarghi-nv
link pred
6266e45
@alexbarghi-nv
Merge branch 'branch-24.10' of https://github.com/rapidsai/cugraph in…
25861ab 
…to pyg-linkpred
@alexbarghi-nv
bug fixes
8a07a0b
@alexbarghi-nv
tests
1433d17
@alexbarghi-nv
link pred single gpu
52bdcc6
@alexbarghi-nv
mnmg
a5ae01e
@alexbarghi-nv
Merge branch 'branch-24.10' into pyg-linkpred
cc2db44
@alexbarghi-nv
fix minor bug
a36899a
@alexbarghi-nv
Merge branch 'branch-24.10' of https://github.com/rapidsai/cugraph in…
f3d6c9b 
…to pyg-linkpred
@alexbarghi-nv
remove batch size 1 due to CI issue
ac410a2
@alexbarghi-nv
Merge branch 'pyg-linkpred' of https://github.com/alexbarghi-nv/cugraph…
4cae976 
… into pyg-linkpred
@alexbarghi-nv
skip snmg example
e413c8a
@alexbarghi-nv
Merge branch 'branch-24.10' into pyg-linkpred
c16a669
@alexbarghi-nv
support negative sampling, coo in PLC
61fb2d3
@alexbarghi-nv alexbarghi-nv self-assigned this Sep 20, 2024
@alexbarghi-nv alexbarghi-nv added this to the 24.10 milestone Sep 24, 2024
@alexbarghi-nv alexbarghi-nv marked this pull request as ready for review September 24, 2024 23:50
@alexbarghi-nv alexbarghi-nv requested review from a team as code owners September 24, 2024 23:50
@alexbarghi-nv alexbarghi-nv requested a review from a team as a code owner September 24, 2024 23:50
ChuckHastings
ChuckHastings approved these changes Sep 25, 2024
View reviewed changes
Copy link
Collaborator

@ChuckHastings ChuckHastings left a comment

Choose a reason for hiding this comment

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

Approving C API updates.

rlratzel
rlratzel approved these changes Sep 27, 2024
View reviewed changes
Copy link
Contributor

@rlratzel rlratzel left a comment

Choose a reason for hiding this comment

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

I'm not familiar with this code and the PR is quite large, but I didn't see anything that jumped out as a problem. I did have one question/suggestion though which need not hold up approval.

python/cugraph-pyg/cugraph_pyg/examples/rgcn_link_class_mnmg.py
Comment on lines +47 to +49
# Ensures that a CUDA context is not created on import of rapids.
# Allows pytorch to create the context instead
os.environ["RAPIDS_NO_INITIALIZE"] = "1"
Copy link
Contributor

Choose a reason for hiding this comment

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

I'm assuming you tested this but I'm surprised this works since it comes after various cugraph imports above. Would it be safer to move up, in case something in cugraph is refactored to pull in a module that creates a CUDA context?

Copy link
Member Author

@alexbarghi-nv alexbarghi-nv Sep 27, 2024 •
edited
Loading

Choose a reason for hiding this comment

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

You're right, this got pulled in from a similar file and ended up in the wrong place. I'll change it.

@alexbarghi-nv
Copy link
Member Author

I'm not familiar with this code and the PR is quite large, but I didn't see anything that jumped out as a problem. I did have one question/suggestion though which need not hold up approval.

This PR includes the other PRs (in-memory sampling, link prediction) which is why there are a lot of changes. But if you've reviewed everything we could merge this in place of the other ones.

@alexbarghi-nv
Copy link
Member Author

/merge

@rapids-bot rapids-bot bot merged commit 4dfd3b3 into rapidsai:branch-24.10 Sep 30, 2024
130 of 131 checks passed
This was referenced Sep 30, 2024
Closed
Closed
Closed
Closed
@alexbarghi-nv alexbarghi-nv deleted the pyg-neg-sampling branch September 30, 2024 15:56
@alexbarghi-nv alexbarghi-nv mentioned this pull request Sep 30, 2024
Merged
alexbarghi-nv added a commit to rapidsai/cugraph-gnn that referenced this pull request Sep 30, 2024
@alexbarghi-nv
Merge pull request #48 from alexbarghi-nv/pyg-neg-sampling
2b6f2cd 
Buffered in-memory sampling in cuGraph-PyG, link prediction in cuGraph-PyG, and negative sampling in cuGraph-PyG (rapidsai/cugraph#4660)
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Reviewers

@ChuckHastings ChuckHastings ChuckHastings approved these changes

@rlratzel rlratzel rlratzel approved these changes

@tingyu66 tingyu66 Awaiting requested review from tingyu66

Assignees

@alexbarghi-nv alexbarghi-nv

Labels
CMake cuGraph feature request New feature or request non-breaking Non-breaking change python
Projects
None yet
Milestone
24.10
Development

Successfully merging this pull request may close these issues.

[FEA] Support Negative Sampling in cuGraph-PyG
3 participants
@alexbarghi-nv @rlratzel @ChuckHastings