-
I see the following warning while precompiling the package:
```
WARNING: Method definition (::GraphNeuralNetworks.GINConv{R
-
Given example:
```julia
using Flux, GraphNeuralNetworks, CUDA
data = [[2,3], [1,4,5], [1], [2,5], [2,4]]
g = GNNGraph(data)
g.num_nodes # 5
g.num_edges # 10
g.num_graphs # 1
s = [1,1…
-
Hello, just found your sweet package!
I ran into a minor issue when wrapping a SimpleWeightedDiGraph with a GNNGraph. The weights from the DiGraph are not included in the wrapped graph.
Below is …
-
As discussed in the closed PR https://github.com/CarloLucibello/GraphNeuralNetworks.jl/pull/376, I investigated the reason why `MLUtils.unbatch` did not work on the GPU. I found out that it uses `geto…
-
**Describe the bug 🐞**
I'm trying to train a Graph NODE integrating `GraphNeuralNetworks.jl` with `OrdinaryDiffEq.jl` where I am learning not only the weights of the neural network, but also the ed…
-
The following snippet causes scalar indexing error on GPU, which is reproducible on the latest release and the master branch.
```julia
using CUDA
using Flux
using GraphNeuralNetworks
CUDA.allow…
-
Hi, I am still running into the error, even after fixing the Windows issue. Could it be that there is an error with accessing the folder before it has even been created? This error only occurs for thi…
-
From the definition of [`AGNNConv`](https://carlolucibello.github.io/GraphNeuralNetworks.jl/stable/api/conv/#GraphNeuralNetworks.AGNNConv), I do not expect the following behaviour:
```julia
using …
-
I think I noticed a bug when adding new edges (with new nodes) into GNNHeteroGraph.
Documentation for `add_edges` says:
> If the edge type is not already present in the graph, it is added. If …
-
```julia
julia> g = rand_graph(3, 2)
GNNGraph:
num_nodes: 3
num_edges: 2
julia> g.ndata.x = rand(1, 2)
ERROR: AssertionError: expected (numobs(x) == getn(ds)) but got 2 != 3
Stacktrace:
…