Improve support in conv layers for edge features (#130)

* add features table for conv layer; use GCConv also without edge features * edge features support in GATv2Conv
JuliaGraphs · Feb 13, 2022 · 00909d4 · 00909d4
1 parent af8ebc1
commit 00909d4
Show file tree

Hide file tree

Showing 4 changed files with 203 additions and 90 deletions.
diff --git a/docs/src/api/conv.md b/docs/src/api/conv.md
@@ -4,12 +4,33 @@ CurrentModule = GraphNeuralNetworks
 
 # Convolutional Layers
 
-## Index 
+Many different types of graphs convolutional layers have been proposed in the literature.
+Choosing the right layer for your application can be a matter of trial and error. 
+Some of the most commonly used layers are the [`GCNConv`](@ref) and the [`GATv2Conv`](@ref) layers. Multiple graph convolutional layers are stacked to create a graph neural network model
+(see [`GNNChain`](@ref)).
+
+The table below lists all graph convolutional layers implemented in the *GraphNeuralNetworks.jl*. It also highlights the presence of some additional capabilities with respect to basic message passing:
+- *Sparse Ops*: implements message passing as multiplication by sparse adjacency matrix instead of the gather/scatter mechanism. This can lead to better cpu performances but it is not supported on gpu yet. 
+- *Edge Weights*: supports scalar weights (or equivalently scalar features) on edges. 
+- *Edge Features*: supports feature vectors on edges.
+
+| Layer                       |Sparse Ops|Edge Weight|Edge Features| 
+| :--------                   |  :---:   |:---:      |:---:        |               
+| [`AGNNConv`](@ref)          |          |           |    ✓        |
+| [`CGConv`](@ref)            |          |           |             |
+| [`ChebConv`](@ref)          |          |           |             |
+| [`EdgeConv`](@ref)          |          |           |             |
+| [`GATConv`](@ref)           |          |           |     ✓       |
+| [`GATv2Conv`](@ref)         |          |           |     ✓       |
+| [`GatedGraphConv`](@ref)    |     ✓    |           |             |
+| [`GCNConv`](@ref)           |     ✓    |     ✓     |             | 
+| [`GINConv`](@ref)           |     ✓    |           |             |
+| [`GraphConv`](@ref)         |     ✓    |           |             |
+| [`MEGNetConv`](@ref)        |          |           |     ✓       |
+| [`NNConv`](@ref)            |          |           |     ✓       |
+| [`ResGatedGraphConv`](@ref) |          |           |             |
+| [`SAGEConv`](@ref)          |     ✓    |           |             |
 
-```@index
-Order = [:type, :function]
-Pages = ["conv.md"]
-```
 
 ## Docs
 

diff --git a/src/deprecations.jl b/src/deprecations.jl
@@ -28,3 +28,6 @@ end
 ## Deprecated in v0.3 
 
 @deprecate copyxj(xi, xj, e)  copy_xj(xi, xj, e) 
+
+@deprecate CGConv(nin::Int, ein::Int, out::Int, args...; kws...) CGConv((nin, ein) => out, args...; kws...)
+
Original file line number	Diff line number	Diff line change
Expand Up		@@ -28,3 +28,6 @@ end
		## Deprecated in v0.3

		@deprecate copyxj(xi, xj, e) copy_xj(xi, xj, e)

		@deprecate CGConv(nin::Int, ein::Int, out::Int, args...; kws...) CGConv((nin, ein) => out, args...; kws...)