🦉 https://github.com/FluxML/Zygote.jl 자료를 번역하는 곳입니당

Zygote

Welcome! Zygote extends the Julia language to support differentiable programming. With Zygote you can write down any Julia code you feel like – including using existing Julia packages – then get gradients and optimise your program. Deep learning, ML and probabilistic programming are all different kinds of differentiable programming that you can do with Zygote.

At least, that's the idea. We're still in beta so expect some adventures.

Setup

Zygote is still moving quickly and it's best to work from the development branches. Run this in a Julia session:

using Pkg; pkg"add Zygote#master"

Taking Gradients

Zygote is easy to understand since, at its core, it has a one-function API (forward), along with a few simple conveniences. Before explaining forward, we'll look at the higher-level function gradient.

gradient calculates derivatives. For example, the derivative of $3x^2 + 2x + 1$ is $6x + 2$, so when x = 5, dx = 32.

julia> using Zygote

julia> gradient(x -> 3x^2 + 2x + 1, 5)
(32,)

gradient returns a tuple, with a gradient for each argument to the function.

julia> gradient((a, b) -> a*b, 2, 3)
(3, 2)

This will work equally well if the arguments are arrays, structs, or any other Julia type, but the function should return a scalar (like a loss or objective $l$, if you're doing optimisation / ML).

julia> W = rand(2, 3); x = rand(3);

julia> gradient(W -> sum(W*x), W)[1]
2×3 Array{Float64,2}:
 0.0462002  0.817608  0.979036
 0.0462002  0.817608  0.979036

julia> gradient(x -> 3x^2 + 2x + 1, 1//4)
(7//2,)

Control flow is fully supported, including recursion.

julia> function pow(x, n)
         r = 1
         for i = 1:n
           r *= x
         end
         return r
       end
pow (generic function with 1 method)

julia> gradient(x -> pow(x, 3), 5)
(75,)

julia> pow2(x, n) = n <= 0 ? 1 : x*pow2(x, n-1)
pow2 (generic function with 1 method)

julia> gradient(x -> pow2(x, 3), 5)
(75,)

Data structures are also supported, including mutable ones like dictionaries. Arrays are currently immutable, though this may change in future.

julia> d = Dict()
Dict{Any,Any} with 0 entries

julia> gradient(5) do x
         d[:x] = x
         d[:x] * d[:x]
       end
(10,)

julia> d[:x]
5

Structs and Types

Julia makes it easy to work with custom types, and Zygote makes it easy to differentiate them. For example, given a simple Point type:

import Base: +, -

struct Point
  x::Float64
  y::Float64
end

a::Point + b::Point = Point(a.x + b.x, a.y + b.y)
a::Point - b::Point = Point(a.x - b.x, a.y - b.y)
dist(p::Point) = sqrt(p.x^2 + p.y^2)

julia> a = Point(1, 2)
Point(1.0, 2.0)

julia> b = Point(3, 4)
Point(3.0, 4.0)

julia> dist(a + b)
7.211102550927978

julia> gradient(a -> dist(a + b), a)[1]
(x = 0.5547001962252291, y = 0.8320502943378437)

Zygote's default representation of the "point adjoint" is a named tuple with gradients for both fields, but this can of course be customised too.

This means we can do something very powerful: differentiating through Julia libraries, even if they weren't designed for this. For example, colordiff might be a smarter loss function on colours than simple mean-squared-error:

julia> using Colors

julia> colordiff(RGB(1, 0, 0), RGB(0, 1, 0))
86.60823557376344

julia> gradient(colordiff, RGB(1, 0, 0), RGB(0, 1, 0))
((r = 0.4590887719632896, g = -9.598786801605689, b = 14.181383399012862), (r = -1.7697549557037275, g = 28.88472330558805, b = -0.044793892637761346))

Gradients of ML models

It's easy to work with even very large and complex models, and there are few ways to do this. Autograd-style models pass around a collection of weights.

julia> linear(θ, x) = θ[:W] * x .+ θ[:b]
linear (generic function with 1 method)

julia> x = rand(5);

julia> θ = Dict(:W => rand(2, 5), :b => rand(2))
Dict{Any,Any} with 2 entries:
  :b => [0.0430585, 0.530201]
  :W => [0.923268 … 0.589691]

# Alternatively, use a named tuple or struct rather than a dict.
# θ = (W = rand(2, 5), b = rand(2))

julia> θ̄ = gradient(θ -> sum(linear(θ, x)), θ)[1]
Dict{Any,Any} with 2 entries:
  :b => [1.0, 1.0]
  :W => [0.628998 … 0.433006]

An extension of this is the Flux-style model in which we use call overloading to combine the weight object with the forward pass (equivalent to a closure).

julia> struct Linear
         W
         b
       end

julia> (l::Linear)(x) = l.W * x .+ l.b

julia> model = Linear(rand(2, 5), rand(2))
Linear([0.267663 … 0.334385], [0.0386873, 0.0203294])

julia> dmodel = gradient(model -> sum(model(x)), model)[1]
(W = [0.652543 … 0.683588], b = [1.0, 1.0])

Zygote also support one more way to take gradients, via implicit parameters – this is a lot like autograd-style gradients, except we don't have to thread the parameter collection through all our code.

julia> W = rand(2, 5); b = rand(2);

julia> linear(x) = W * x .+ b
linear (generic function with 2 methods)

julia> grads = gradient(() -> sum(linear(x)), Params([W, b]))
Grads(...)

julia> grads[W], grads[b]
([0.652543 … 0.683588], [1.0, 1.0])

However, implicit parameters exist mainly for compatibility with Flux's current AD; it's recommended to use the other approaches unless you need this.