Open realCrush opened 2 years ago
I think this is a great suggestion :) If you do not have any plans to look into this, I can try to add this dataset as well.
I think this is a great suggestion :) If you do not have any plans to look into this, I can try to add this dataset as well.
sure, please have a try🥳
I think this is a great suggestion :) If you do not have any plans to look into this, I can try to add this dataset as well.
By the way, I would be super appreciate if you could provide some model training examples on VarMisuse with PyG models, such as GNN-FiLM!😋
If no one is working on this, can I take it up?
If no one is working on this, can I take it up?
Sure! Just do it.
@Cyberpuncrush I have implemented it but some things need to be done
https://github.com/pyg-team/pytorch_geometric/pull/3978
Let me know if I am on right track.
@arvindmuralie77 You can improve on this implementation if you wish.
Any update on this dataset? Long range problems on graphs is hot in recent days, so this real and large scale dataset would be great to probe the performance of GNNs.
The PR slipped under my radar, let me try to revisit it. Sorry for any delay.
Is it still in progress?
🚀 Feature
Motivation
paper "LEARNING TO REPRESENT PROGRAMS WITH GRAPHS" which encode computer programs as graphs, with rich semantic information, however, most code implementation on this dataset VarMisuse is based on TensorFlow, like tf-gnn-samples
Pitch
include VarMisuse dataset in PyG framework as a class under torch_geometric.datasets
Additional context
raw dataset provided by author: VarMisuse
more papers/models on this dataset: GNN-FiLM: Graph Neural Networks with Feature-wise Linear Modulation ON THE BOTTLENECK OF GRAPH NEURAL NETWORKS AND ITS PRACTICAL IMPLICATIONS