Try #728:

.gitlab-ci.yml

@@ -5,7 +5,7 @@ image: nvidia/cuda:10.1-cudnn7-devel-ubuntu18.04
 
 julia:1.3:
   extends:
-    - .julia:1.3
+    - .julia:1.4
     - .test
   tags:
     - nvidia
@@ -20,7 +20,7 @@ julia:nightly:
 
 documentation:
   extends:
-    - .julia:1.3
+    - .julia:1.4
     - .documentation
   tags:
     - nvidia

Project.toml

@@ -13,6 +13,7 @@ Future = "9fa8497b-333b-5362-9e8d-4d0656e87820"
 IRTools = "7869d1d1-7146-5819-86e3-90919afe41df"
 InteractiveUtils = "b77e0a4c-d291-57a0-90e8-8db25a27a240"
 LinearAlgebra = "37e2e46d-f89d-539d-b4ee-838fcccc9c8e"
+LoopVectorization = "bdcacae8-1622-11e9-2a5c-532679323890"
 MacroTools = "1914dd2f-81c6-5fcd-8719-6d5c9610ff09"
 NNlib = "872c559c-99b0-510c-b3b7-b6c96a88d5cd"
 NaNMath = "77ba4419-2d1f-58cd-9bb1-8ffee604a2e3"
@@ -36,7 +37,7 @@ NaNMath = "0.3"
 Requires = "0.5, 1.0"
 SpecialFunctions = "0.10"
 ZygoteRules = "0.2"
-julia = "1.3"
+julia = "1.4"
 
 [extras]
 CUDA = "052768ef-5323-5732-b1bb-66c8b64840ba"

src/Zygote.jl

@@ -11,6 +11,8 @@ using IRTools
 using MacroTools, Requires
 using MacroTools: @forward
 
+using LoopVectorization # for vmap
+
 export Params, gradient, pullback, pushforward, @code_grad
 
 include("tools/idset.jl")

src/lib/array.jl

@@ -170,23 +170,25 @@ function unzip(tuples)
   _unzip(tuples, Val(N))
 end
 
-function ∇map(cx, f, args...)
-  ys_and_backs = map((args...) -> _pullback(cx, f, args...), args...)
-  if isempty(ys_and_backs)
-    ys_and_backs, _ -> nothing
-  else
-    ys, backs = unzip(ys_and_backs)
-    ys, function (Δ)
-      Δf_and_args_zipped = map((f, δ) -> f(δ), backs, Δ)
-      Δf_and_args = unzip(Δf_and_args_zipped)
-      Δf = reduce(accum, Δf_and_args[1])
-      (Δf, Δf_and_args[2:end]...)
+for (mapfunc,∇mapfunc) in [(:map,:∇map),(:pmap,:∇pmap),(:vmap,:∇vmap)]
+  @eval function $∇mapfunc(cx, f, args...)
+    ys_and_backs = $mapfunc((args...) -> _pullback(cx, f, args...), args...)
+    if isempty(ys_and_backs)
+      ys_and_backs, _ -> nothing
+    else
+      ys, backs = unzip(ys_and_backs)
+      ys, function (Δ)
+        Δf_and_args_zipped = $mapfunc((f, δ) -> f(δ), backs, Δ)
+        Δf_and_args = unzip(Δf_and_args_zipped)
+        Δf = reduce(accum, Δf_and_args[1])
+        (Δf, Δf_and_args[2:end]...)
+      end
     end
   end
-end
 
-@adjoint function map(f, args::Union{AbstractArray,Tuple}...)
-  ∇map(__context__, f, args...)
+  @eval @adjoint function $mapfunc(f, args::Union{AbstractArray,Tuple}...)
+    $∇mapfunc(__context__, f, args...)
+  end
 end
 
 function _pullback(cx::AContext, ::typeof(collect), g::Base.Generator)

test/gradcheck.jl

@@ -3,6 +3,7 @@ using Zygote, NNlib, Test, Random, LinearAlgebra, Statistics, FillArrays,
 using Zygote: gradient
 using NNlib: conv, ∇conv_data, depthwiseconv, batched_mul
 using Base.Broadcast: broadcast_shape
+using LoopVectorization
 
 function ngradient(f, xs::AbstractArray...)
   grads = zero.(xs)
@@ -103,7 +104,7 @@ end
 @test gradtest((w, x) -> parent(w)*x, randn(5,5)', randn(5,5))
 @test gradtest((w, x) -> parent(w)*x, transpose(randn(5,5)), randn(5,5))
 
-@testset "sum, prod, cumsum" begin 
+@testset "sum, prod, cumsum" begin
   @test gradtest(x -> sum(x, dims = (2, 3)), (3,4,5))
   @test gradtest(x -> sum(abs2, x), randn(4, 3, 2))
   @test gradtest(x -> sum(abs2, x; dims=1), randn(4, 3, 2))
@@ -301,15 +302,17 @@ end
 @test gradtest(kron, rand(5,1), rand(3,1), rand(8,1))
 @test gradtest(kron, rand(5,2), rand(3,2), rand(8,2))
 
-@testset "map" begin
-  @test gradtest(xs -> sum(map(x -> x^2, xs)), rand(2,3))
-  @test gradtest((xss...) -> sum(map((xs...) -> sqrt(sum(xs.^2)), xss...)), [rand(5) for _ in 1:6]...)
-  function foo(y)
-    bar = (x) -> x*y
-    sum(map(bar, 1:5))
+for mapfunc in [map,pmap,vmap]
+  @testset "$mapfunc" begin
+    @test gradtest(xs -> sum(mapfunc(x -> x^2, xs)), rand(2,3))
+    @test gradtest((xss...) -> sum(mapfunc((xs...) -> sqrt(sum(xs.^2)), xss...)), [rand(5) for _ in 1:6]...)
+    function foo(y)
+      bar = (x) -> x*y
+      sum(mapfunc(bar, 1:5))
+    end
+    @test gradtest(foo, 3)
+    @test gradient(v -> sum([x for x in v]), [1.1,2.2,3.3]) == ([1, 1, 1],)
   end
-  @test gradtest(foo, 3)
-  @test gradient(v -> sum([x for x in v]), [1.1,2.2,3.3]) == ([1, 1, 1],)
 end
 
 @testset "sort" begin