fix lstm

Author: CarloLucibello

GNNLux/Project.toml

@@ -1,7 +1,7 @@
 name = "GNNLux"
 uuid = "e8545f4d-a905-48ac-a8c4-ca114b98986d"
 authors = ["Carlo Lucibello and contributors"]
-version = "0.1.0"
+version = "0.2.0-DEV"
 
 [deps]
 ConcreteStructs = "2569d6c7-a4a2-43d3-a901-331e8e4be471"
@@ -18,7 +18,7 @@ Statistics = "10745b16-79ce-11e8-11f9-7d13ad32a3b2"
 [compat]
 ConcreteStructs = "0.2.3"
 GNNGraphs = "1.3"
-GNNlib = "0.2.3"
+GNNlib = "1"
 Lux = "1"
 LuxCore = "1"
 NNlib = "0.9.21"

GNNlib/Project.toml

@@ -1,7 +1,7 @@
 name = "GNNlib"
 uuid = "a6a84749-d869-43f8-aacc-be26a1996e48"
 authors = ["Carlo Lucibello and contributors"]
-version = "0.2.3"
+version = "1.0.0-DEV"
 
 [deps]
 ChainRulesCore = "d360d2e6-b24c-11e9-a2a3-2a2ae2dbcce4"

GNNlib/src/layers/pool.jl

@@ -29,8 +29,11 @@ topk_index(y::Adjoint, k::Int) = topk_index(y', k)
 function set2set_pool(l, g::GNNGraph, x::AbstractMatrix)
     n_in = size(x, 1)    
     qstar = zeros_like(x, (2*n_in, g.num_graphs))
+    h = zeros_like(l.lstm.Wh, size(l.lstm.Wh, 2))
+    c = zeros_like(l.lstm.Wh, size(l.lstm.Wh, 2))
     for t in 1:l.num_iters
-        q = l.lstm(qstar)                            # [n_in, n_graphs]
+        h, c = l.lstm(qstar, (h, c))                     # [n_in, n_graphs]
+        q = h
         qn = broadcast_nodes(g, q)                    # [n_in, n_nodes]
         α = softmax_nodes(g, sum(qn .* x, dims = 1))  # [1, n_nodes]
         r = reduce_nodes(+, g, x .* α)               # [n_in, n_graphs]

GraphNeuralNetworks/Project.toml

@@ -19,8 +19,8 @@ Statistics = "10745b16-79ce-11e8-11f9-7d13ad32a3b2"
 [compat]
 ChainRulesCore = "1"
 Flux = "0.15"
-GNNGraphs = "1.0"
-GNNlib = "0.2"
+GNNGraphs = "1"
+GNNlib = "1"
 LinearAlgebra = "1"
 MLUtils = "0.4"
 MacroTools = "0.5"

GraphNeuralNetworks/src/layers/pool.jl

@@ -149,24 +149,19 @@ end
 Flux.@layer Set2Set
 
 function Set2Set(n_in::Int, n_iters::Int, n_layers::Int = 1)
-    @assert n_layers >= 1
+    @assert n_layers == 1 "multiple layers not implemented yet" #TODO
     n_out = 2 * n_in
-
-    if n_layers == 1
-        lstm = LSTM(n_out => n_in)
-    else
-        layers = [LSTM(n_out => n_in)]
-        for _ in 2:n_layers
-            push!(layers, LSTM(n_in => n_in))
-        end
-        lstm = Chain(layers...)
-    end
-
+    lstm = LSTMCell(n_out => n_in)
     return Set2Set(lstm, n_iters)
 end
 
+function initialstates(cell::LSTMCell)
+    h = zeros_like(cell.Wh, size(cell.Wh, 2))
+    c = zeros_like(cell.Wh, size(cell.Wh, 2))
+    return h, c
+end
+
 function (l::Set2Set)(g, x)
-    Flux.reset!(l.lstm)
     return GNNlib.set2set_pool(l, g, x)
 end
 

GraphNeuralNetworks/test/layers/pool.jl

@@ -76,7 +76,7 @@ end
 
         n_in = 3
         n_iters = 2
-        n_layers = 1
+        n_layers = 1 #TODO test with more layers
         g = batch([rand_graph(10, 40, graph_type = GRAPH_T) for _ in 1:5])
         g = GNNGraph(g, ndata = rand(Float32, n_in, g.num_nodes))
         l = Set2Set(n_in, n_iters, n_layers)