Deeptendu Santra
Deeptendu Santra
Hello devs. I am trying to develop support for OGB Datasets in [MLDatasets.jl](https://github.com/JuliaML/MLDatasets.jl). One of the bottlenecks we are facing is loading the .pt files. This implementation [here](https://github.com/JuliaML/MLDatasets.jl/pull/172) using Pickle.jl...
All Graph Dataset from PyTorch Geometric Till date (29/09/2022) - [x] KarateClub - [x] TUDataset - [ ] GNNBenchMarkDataset - [x] Planetoid - [ ] FakeDataset - [ ] FakeHeteroDataset...
Related to: #169 Added the GraphDataset struct which can replace all custom structs for each Data source. The metadata will have fields like OGBDatasets metadata, including name, source, etc. We...
It seems that support for non-large Stanford OGB Datasets is almost complete except for these few issues: - [ ] Split for OGBDataset involving HeteroGraph with link task - [...
It would be good to have something like: ```julia struct GraphDataset
Tutorial on Graph Segmentation using FAUST inspired from this [medium article](https://medium.com/stanford-cs224w/deep-learning-on-3d-meshes-9608a5b33c98).
There are cases where the shape of the feature array might not have the last dim = number of nodes and edges. For example: In a mesh represented by a...
Currently, GraphNeuralNetworks.jl is added as a local package in [notebooks](https://github.com/CarloLucibello/GraphNeuralNetworks.jl/blob/master/docs/src/tutorials/graph_classification_pluto.jl#L17) . This can hinder users from running the notebook just after downloading. So, we should use the upstream GraphNeuralNetworks.jl for...