-
Notifications
You must be signed in to change notification settings - Fork 3.7k
Issues: pyg-team/pytorch_geometric
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
Author
Label
Projects
Milestones
Assignee
Sort
Issues list
Normalization change in
AddRandomWalkPE
: From A D⁻¹ to D⁻¹ May Deviate from Paper
bug
#9915
opened Jan 3, 2025 by
ywangmy
GlobalStorage#is_node_attr() breaks when cat_dim is a tuple
bug
#9895
opened Dec 26, 2024 by
Wollents
After call NeighborLoader my file logger prints in console
bug
loader
#9892
opened Dec 26, 2024 by
munkhbuyan
Running on multiple GPUs is slower than on a single GPU
distributed
question
#9891
opened Dec 26, 2024 by
HanChen-HUST
Scatter leads to wrong results on gpu with cast to short()-tensor
bug
#9886
opened Dec 23, 2024 by
maichmueller
Integrating GNNs and LLMs for Enhanced Data Warehouse Understanding and Lineage Analysis
feature
#9839
opened Dec 10, 2024 by
AJamal27891
Implementing Bag of Tricks for Node Classification with Graph Neural Networks
feature
#9831
opened Dec 9, 2024 by
mattjhayes3
Memory leak in GraphUNet due to function to_torch_csr_tensor
bug
#9826
opened Dec 6, 2024 by
acotino-ignitioncomputing
ValueError("Expected a parent")
when using new Lightning imports and PyG Lightning DataModule wrappers
bug
#9811
opened Nov 27, 2024 by
nathanpainchaud
Optional packages torch-sparse, torch-cluster, and pyg-lib are missing dependencies
installation
#9809
opened Nov 27, 2024 by
ErasonElisabeth
conda cannot install latest version (2.6.1) with pytorch 2.5.*
installation
#9795
opened Nov 16, 2024 by
themaigod
Install a full PyG environment using only a single pip command
installation
#9792
opened Nov 15, 2024 by
Anjum48
Previous Next
ProTip!
Find all open issues with in progress development work with linked:pr.