support maps over tuples

src/backend/mxnet/graph.jl

@@ -40,6 +40,8 @@ graph(::typeof(softmax), xs) =
 graph(::typeof(cat), dim::Integer, a...) = mx.Concat(a..., dim = dim)
 graph(::typeof(vcat), a...) = graph(cat, 1, a...)
 
+graph(::typeof(map), f, xss::Tuple...) = map(f, xss...)
+
 graph(::Input, x) = x
 
 graph(ctx::Context, d::Affine, x) =

src/backend/tensorflow/graph.jl

@@ -39,6 +39,8 @@ end
 
 graph(::typeof(.-), args...) = -(args...)
 
+graph(::typeof(map), f, xss::Tuple...) = map(f, xss...)
+
 # reshape hack due to https://github.com/malmaud/TensorFlow.jl/issues/79
 batchsize(x::Tensor) = reduce_sum(slice(TensorFlow.shape(x), [0], [1]))
 graph(::typeof(flatten), x) = reshape(x, pack([batchsize(x), Int32(-1)]))

test/backend/common.jl

@@ -55,3 +55,11 @@ function test_stacktrace(bk)
     @test e.trace[2].func == :TLP
   end
 end
+
+function test_anon(bk)
+  @testset "Closures" begin
+    x, y = rand(3), rand(5)
+    model = bk(@net xs -> map(x -> x .* x, xs))
+    @test all(model((x, y)) .≈ (x.*x, y.*y))
+  end
+end

test/backend/mxnet.jl

@@ -13,6 +13,7 @@ test_tupleio(mxnet)
 test_recurrence(mxnet)
 test_stacktrace(mxnet)
 test_back(mxnet)
+test_anon(mxnet)
 
 @testset "Native interface" begin
   f = mx.FeedForward(Chain(d, softmax))

test/backend/tensorflow.jl

@@ -12,6 +12,7 @@ dt = tf(d)
 test_tupleio(tf)
 test_recurrence(tf)
 test_stacktrace(tf)
+test_anon(tf)
 
 @testset "Tensor interface" begin
   sess = TensorFlow.Session()

test/basic.jl

@@ -18,6 +18,8 @@ Flux.infer(d, (1, 10))
 #   @test isa(x, DataFlow.Input) && isa(W, Param) && isa(b, Param)
 # end
 
+test_anon(identity)
+
 let a1 = Affine(10, 20), a2 = Affine(20, 15)
   tlp = TLP(a1, a2)
   @test tlp(xs) ≈ softmax(a2(σ(a1(xs))))