more docs

Author: CarloLucibello

docs/Project.toml

@@ -1,3 +1,5 @@
 [deps]
 Documenter = "e30172f5-a6a5-5a46-863b-614d45cd2de4"
+Flux = "587475ba-b771-5e3f-ad9e-33799f191a9c"
 GraphNeuralNetworks = "cffab07f-9bc2-4db1-8861-388f63bf7694"
+NNlib = "872c559c-99b0-510c-b3b7-b6c96a88d5cd"

docs/make.jl

@@ -1,4 +1,4 @@
-using GraphNeuralNetworks
+using Flux, NNlib, GraphNeuralNetworks
 using Documenter
 
 DocMeta.setdocmeta!(GraphNeuralNetworks, :DocTestSetup, :(using GraphNeuralNetworks); recursive=true)

docs/src/api/messagepassing.md

@@ -0,0 +1,7 @@
+# Message Passing
+
+```@docs
+GraphNeuralNetworks.message
+GraphNeuralNetworks.update
+GraphNeuralNetworks.propagate
+```

docs/src/messagepassing.md

@@ -1,7 +1,2 @@
 # Message Passing
 
-```@docs
-message
-update
-propagate
-```

docs/src/models.md

@@ -6,7 +6,7 @@ their models.
 
 In what follows, we discuss two different styles for model creation:
 the *explicit modeling* style, more verbose but more flexible, 
-and the *implicity modeling* style based on [`GNNChain`](@ref), more concise but less flexible.
+and the *implicit modeling* style based on [`GNNChain`](@ref), more concise but less flexible.
 
 ## Explicit modeling
 

src/gnngraph.jl

@@ -26,7 +26,7 @@ is preserved and shared.
 A `GNNGraph` is a LightGraphs' `AbstractGraph`, therefore any functionality
 from the LightGraphs' graph library can be used on it.
 
-## Arguments 
+# Arguments 
 
 - `data`: Some data representing the graph topology. Possible type are 
     - An adjacency matrix
@@ -301,7 +301,7 @@ end
 
 Normalized Laplacian matrix of graph `g`.
 
-## Arguments
+# Arguments
 
 - `g`: A `GNNGraph`.
 - `T`: result element type.
@@ -331,7 +331,7 @@ end
 Scaled Laplacian matrix of graph `g`,
 defined as ``\hat{L} = \frac{2}{\lambda_{max}} L - I`` where ``L`` is the normalized Laplacian matrix.
 
-## Arguments
+# Arguments
 
 - `g`: A `GNNGraph`.
 - `T`: result element type.

src/layers/basic.jl

@@ -17,7 +17,7 @@ and if names are given, `m[:name] == m[1]` etc.
 
 ## Examples
 
-```jldoctest
+```
 julia> m = GNNChain(x -> x^2, x -> x+1);
 
 julia> m(5) == 26

src/layers/conv.jl

@@ -12,7 +12,7 @@ where ``c_{ij} = \sqrt{N(i)\,N(j)}``.
 The input to the layer is a node feature array `X` 
 of size `(num_features, num_nodes)`.
 
-## Arguments
+# Arguments
 
 - `in`: Number of input features.
 - `out`: Number of output features.
@@ -89,7 +89,7 @@ Z^{(k)} = 2 \hat{L} Z^{(k-1)} - Z^{(k-2)}
 
 with ``\hat{L}`` the [`scaled_laplacian`](@ref).
 
-## Arguments
+# Arguments
 
 - `in`: The dimension of input features.
 - `out`: The dimension of output features.
@@ -151,7 +151,7 @@ Performs:
 ```
 where the aggregation type is selected by `aggr`.
 
-## Arguments
+# Arguments
 
 - `in`: The dimension of input features.
 - `out`: The dimension of output features.
@@ -220,7 +220,7 @@ where the attention coefficient ``\alpha_{ij}`` is given by
 ```
 with ``z_i`` a normalization factor.
 
-## Arguments
+# Arguments
 
 - `in`: The dimension of input features.
 - `out`: The dimension of output features.
@@ -306,7 +306,7 @@ Implements the recursion
 
  where ``\mathbf{h}^{(l)}_i`` denotes the ``l``-th hidden variables passing through GRU. The dimension of input ``\mathbf{x}_i`` needs to be less or equal to `out`.
 
-## Arguments
+# Arguments
 
 - `out`: The dimension of output features.
 - `num_layers`: The number of gated recurrent unit.
@@ -374,7 +374,7 @@ Performs the operation
 
 where `f` typically denotes a learnable function, e.g. a linear layer or a multi-layer perceptron.
 
-## Arguments
+# Arguments
 
 - `f`: A (possibly learnable) function acting on edge features. 
 - `aggr`: Aggregation operator for the incoming messages (e.g. `+`, `*`, `max`, `min`, and `mean`).
@@ -418,7 +418,7 @@ Graph Isomorphism convolutional layer from paper [How Powerful are Graph Neural
 ```
 where `f` typically denotes a learnable function, e.g. a linear layer or a multi-layer perceptron.
 
-## Arguments
+# Arguments
 
 - `f`: A (possibly learnable) function acting on node features. 
 - `eps`: Weighting factor.

src/layers/pool.jl

@@ -7,7 +7,7 @@ Global pooling layer.
 
 It pools all features with `aggr` operation.
 
-## Arguments
+# Arguments
 
 - `aggr`: An aggregate function applied to pool all features.
 """
@@ -29,7 +29,7 @@ Local pooling layer.
 
 It pools features with `aggr` operation accroding to `cluster`. It is implemented with `scatter` operation.
 
-## Arguments
+# Arguments
 
 - `aggr`: An aggregate function applied to pool all features.
 - `cluster`: An index structure which indicates what features to aggregate with.
@@ -46,7 +46,7 @@ end
 
 Top-k pooling layer.
 
-## Arguments
+# Arguments
 
 - `adj`: Adjacency matrix  of a graph.
 - `k`: Top-k nodes are selected to pool together.

src/msgpass.jl

@@ -63,7 +63,7 @@ in order to [`update`](@ref) the features of node `i`.
 By default, the function returns `x_j`.
 Custom layer should specialize this method with the desired behavior.
 
-## Arguments
+# Arguments
 
 - `mp`: A gnn layer.
 - `x_i`: Features of the central node `i`.
@@ -86,7 +86,7 @@ aggregation `m̄`.
 By default, the function returns `m̄`.
 Custom layers should  specialize this method with the desired behavior.
 
-## Arguments
+# Arguments
 
 - `mp`: A gnn layer.
 - `m̄`: Aggregated edge messages from the [`message`](@ref) function.