Deeptendu Santra
Deeptendu Santra
### Bug Description Docker cpu image with tag `cpu-9193b92` does not contain the DNNL file. #### Reproduction Steps Create a folder with a `Dockerfile`, `main.cpp` and `Cmakelists.txt` The Dockerfile for...
Closes #155
Currently, the SMSSpam collection cannot be downloaded in CI environments, preventing it from testing. For debug log, check [this run](https://github.com/JuliaML/MLDatasets.jl/runs/7524820232) Removing it from the CI testable datasets until the issue...
## Problems 1. Splitting graphs is a bit more complicated than normal data. Graphs can be split based on node, edge or whole graph itself. We should be able to...
Required for [GraphNeuralNetworks.jl #173](https://github.com/CarloLucibello/GraphNeuralNetworks.jl/issues/173) [Download Link](http://faust.is.tue.mpg.de/challenge/Inter-subject_challenge/datasets). User needs to sign in before downloading the dataset.
Currently MLDatasets.jl downloads Datasets from DataHubs like TUDataset without triggering a question for downloading the dataset and respect `DATADEPS_ALWAYS_ACCEPT`. Opening this issue because the downloading experience is not consistent. If...
Edge indices in undirected HeteroGraph are coded as, ```julia edge_index[triplet] = ([s; t], [t; s]) ``` Here `s` and `t` are node indices of two different types of nodes, so...
Now that we support HeteroGeneous Graphs in [MLDatasets.jl](https://github.com/JuliaML/MLDatasets.jl/pull/140). We should support HeteroGeneous graphs. Ideally, ```julia g = mldataset2gnngraph(heterogeneous_dataset) ```