Merge pull request #591 from dhairyagandhi96/onehot

Fix OneHotVector/Matrix performance on GPU

src/onehot.jl

@@ -9,6 +9,8 @@ Base.size(xs::OneHotVector) = (Int64(xs.of),)
 
 Base.getindex(xs::OneHotVector, i::Integer) = i == xs.ix
 
+Base.getindex(xs::OneHotVector, ::Colon) = xs
+
 A::AbstractMatrix * b::OneHotVector = A[:, b.ix]
 
 struct OneHotMatrix{A<:AbstractVector{OneHotVector}} <: AbstractMatrix{Bool}
@@ -22,6 +24,21 @@ Base.getindex(xs::OneHotMatrix, i::Integer, j::Integer) = xs.data[j][i]
 Base.getindex(xs::OneHotMatrix, ::Colon, i::Integer) = xs.data[i]
 Base.getindex(xs::OneHotMatrix, ::Colon, i::AbstractArray) = OneHotMatrix(xs.height, xs.data[i])
 
+Base.getindex(xs::Flux.OneHotMatrix, j::Base.UnitRange, i::Int) = xs.data[i][j]
+
+Base.getindex(xs::OneHotMatrix, ::Colon, ::Colon) = xs
+
+# handle special case for when we want the entire column
+function Base.getindex(xs::Flux.OneHotMatrix{T}, ot::Union{Base.Slice, Base.OneTo}, i::Int) where {T<:AbstractArray}
+  res = similar(xs, size(xs, 1), 1)
+  if length(ot) == size(xs, 1)
+    res = xs[:,i]
+  else
+    res = xs[1:length(ot),i]
+  end
+  res
+end
+
 A::AbstractMatrix * B::OneHotMatrix = A[:, map(x->x.ix, B.data)]
 
 Base.hcat(x::OneHotVector, xs::OneHotVector...) = OneHotMatrix(length(x), [x, xs...])
@@ -54,13 +71,17 @@ end
 onehotbatch(ls, labels, unk...) =
   OneHotMatrix(length(labels), [onehot(l, labels, unk...) for l in ls])
 
+Base.argmax(xs::OneHotVector) = xs.ix
+
 onecold(y::AbstractVector, labels = 1:length(y)) = labels[Base.argmax(y)]
 
 onecold(y::AbstractMatrix, labels...) =
   dropdims(mapslices(y -> onecold(y, labels...), y, dims=1), dims=1)
 
+onecold(y::OneHotMatrix, labels...) = map(x -> onecold(x, labels...), y.data)
+
 function argmax(xs...)
-  Base.depwarn("`argmax(...) is deprecated, use `onecold(...)` instead.", :argmax)
+  Base.depwarn("`argmax(...)` is deprecated, use `onecold(...)` instead.", :argmax)
   return onecold(xs...)
 end
 

test/cuda/cuda.jl

@@ -38,6 +38,13 @@ Flux.back!(sum(l))
 
 end
 
+@testset "onecold gpu" begin
+  x = zeros(Float32, 10, 3) |> gpu;
+  y = Flux.onehotbatch(ones(3), 1:10) |> gpu;
+  res = Flux.onecold(x) .== Flux.onecold(y)
+  @test res isa CuArray
+end
+
 if CuArrays.libcudnn != nothing
     @info "Testing Flux/CUDNN"
     include("cudnn.jl")