awful hack to get both examples working

src/backend/mxnet/graph.jl

@@ -7,7 +7,7 @@ graph(vars, model, args...) = node(model, args...)
 graph(vars, x::mx.SymbolicNode) = x
 
 # TODO: detect parameters used more than once
-function graph{T<:AArray}(vars, p::Flux.Param{T})
+function graph{T<:AArray}(vars::Associative, p::Flux.Param{T})
   value = p.x
   id = gensym()
   vars[id] = value
@@ -68,7 +68,7 @@ graph(vars, p::MaxPool, x) =
              stride = p.stride)
 
 # TODO: fix the initialisation issue
-graph(vars, d::Dense, x) =
+graph(vars::Void, d::Dense, x) =
   mx.FullyConnected(data = x,
                     num_hidden = size(d.W.x, 1),
                     # weight = graph(vars, d.W),

src/backend/mxnet/model.jl

@@ -47,8 +47,8 @@ function load!(model::MXModel)
   return model
 end
 
-function mxgraph(model, input)
-  vars = Dict{Symbol,Any}()
+function mxgraph(model, input; vars = true)
+  vars = vars ? Dict{Symbol,Any}() : nothing
   node = graph(vars, model, mx.Variable(input))
   return node, vars
 end
@@ -91,6 +91,6 @@ end
 
 function mx.FeedForward(model::Model; input = :data, label = :softmax, context = mx.cpu())
   model = rewrite_softmax(model, label)
-  node, _ = mxgraph(model, input)
+  node, _ = mxgraph(model, input, vars = false)
   return mx.FeedForward(node, context = context)
 end