Julia & Lux for the Uninitiated

This is a quick intro to Lux loosely based on:

PyTorch's tutorial.
Flux's tutorial (the link for which has now been lost to abyss).
Jax's tutorial.

It introduces basic Julia programming, as well Zygote, a source-to-source automatic differentiation (AD) framework in Julia. We'll use these tools to build a very simple neural network. Let's start with importing Lux.jl

julia

using Lux, Random

Precompiling Lux...
    789.2 ms  ✓ ConcreteStructs
    493.4 ms  ✓ SIMDTypes
    514.4 ms  ✓ Reexport
    503.9 ms  ✓ CommonWorldInvalidations
    517.5 ms  ✓ IfElse
    878.6 ms  ✓ OpenLibm_jll
    557.8 ms  ✓ FastClosures
    608.9 ms  ✓ ArgCheck
    682.7 ms  ✓ ConstructionBase
    741.7 ms  ✓ CompilerSupportLibraries_jll
    813.5 ms  ✓ DocStringExtensions
    831.8 ms  ✓ Statistics
    863.5 ms  ✓ EnzymeCore
    520.2 ms  ✓ Future
    550.7 ms  ✓ ManualMemory
    568.4 ms  ✓ StaticArraysCore
    569.8 ms  ✓ HashArrayMappedTries
    677.8 ms  ✓ Requires
    854.8 ms  ✓ ADTypes
    813.1 ms  ✓ Compat
    757.3 ms  ✓ JLLWrappers
    636.6 ms  ✓ ConstructionBase → ConstructionBaseLinearAlgebraExt
    659.6 ms  ✓ NaNMath
    917.6 ms  ✓ CpuId
   1578.4 ms  ✓ IrrationalConstants
    505.3 ms  ✓ ScopedValues
    573.4 ms  ✓ DiffResults
    483.1 ms  ✓ ADTypes → ADTypesConstructionBaseExt
    614.7 ms  ✓ Adapt
   1144.7 ms  ✓ Static
    571.2 ms  ✓ ADTypes → ADTypesEnzymeCoreExt
    579.7 ms  ✓ Compat → CompatLinearAlgebraExt
   1000.7 ms  ✓ ThreadingUtilities
    815.7 ms  ✓ OpenSpecFun_jll
    781.3 ms  ✓ LogExpFunctions
    513.9 ms  ✓ EnzymeCore → AdaptExt
   2426.1 ms  ✓ UnsafeAtomics
    576.5 ms  ✓ BitTwiddlingConvenienceFunctions
    654.6 ms  ✓ GPUArraysCore
    735.8 ms  ✓ ArrayInterface
   2690.3 ms  ✓ MacroTools
    788.2 ms  ✓ Functors
    436.5 ms  ✓ ArrayInterface → ArrayInterfaceGPUArraysCoreExt
    672.8 ms  ✓ Atomix
    506.5 ms  ✓ ArrayInterface → ArrayInterfaceStaticArraysCoreExt
   1323.3 ms  ✓ CPUSummary
    797.3 ms  ✓ CommonSubexpressions
   1521.3 ms  ✓ ChainRulesCore
    897.6 ms  ✓ MLDataDevices
    479.3 ms  ✓ ADTypes → ADTypesChainRulesCoreExt
    818.1 ms  ✓ PolyesterWeave
    511.3 ms  ✓ ArrayInterface → ArrayInterfaceChainRulesCoreExt
   1512.0 ms  ✓ StaticArrayInterface
   1709.1 ms  ✓ DispatchDoctor
    852.0 ms  ✓ MLDataDevices → MLDataDevicesChainRulesCoreExt
   1883.4 ms  ✓ Setfield
    624.4 ms  ✓ CloseOpenIntervals
    638.7 ms  ✓ LayoutPointers
    578.2 ms  ✓ DispatchDoctor → DispatchDoctorEnzymeCoreExt
   1472.8 ms  ✓ LogExpFunctions → LogExpFunctionsChainRulesCoreExt
   1606.1 ms  ✓ Optimisers
    730.8 ms  ✓ DispatchDoctor → DispatchDoctorChainRulesCoreExt
   3132.3 ms  ✓ SpecialFunctions
    434.2 ms  ✓ Optimisers → OptimisersEnzymeCoreExt
    435.4 ms  ✓ Optimisers → OptimisersAdaptExt
    926.3 ms  ✓ StrideArraysCore
   1344.7 ms  ✓ LuxCore
    648.5 ms  ✓ DiffRules
    462.4 ms  ✓ LuxCore → LuxCoreEnzymeCoreExt
    523.5 ms  ✓ LuxCore → LuxCoreMLDataDevicesExt
    572.3 ms  ✓ LuxCore → LuxCoreFunctorsExt
    584.5 ms  ✓ LuxCore → LuxCoreSetfieldExt
    658.7 ms  ✓ LuxCore → LuxCoreChainRulesCoreExt
    810.8 ms  ✓ Polyester
   1713.3 ms  ✓ SpecialFunctions → SpecialFunctionsChainRulesCoreExt
   2872.8 ms  ✓ WeightInitializers
   7327.7 ms  ✓ StaticArrays
    606.3 ms  ✓ Adapt → AdaptStaticArraysExt
    614.3 ms  ✓ StaticArrays → StaticArraysStatisticsExt
    637.1 ms  ✓ ConstructionBase → ConstructionBaseStaticArraysExt
    650.8 ms  ✓ StaticArrays → StaticArraysChainRulesCoreExt
    664.6 ms  ✓ StaticArrayInterface → StaticArrayInterfaceStaticArraysExt
    877.2 ms  ✓ WeightInitializers → WeightInitializersChainRulesCoreExt
   3578.8 ms  ✓ ForwardDiff
    769.2 ms  ✓ ForwardDiff → ForwardDiffStaticArraysExt
   3350.4 ms  ✓ KernelAbstractions
    636.5 ms  ✓ KernelAbstractions → LinearAlgebraExt
    706.9 ms  ✓ KernelAbstractions → EnzymeExt
   5207.9 ms  ✓ NNlib
    816.3 ms  ✓ NNlib → NNlibEnzymeCoreExt
    831.0 ms  ✓ NNlib → NNlibSpecialFunctionsExt
    875.8 ms  ✓ NNlib → NNlibForwardDiffExt
   7501.3 ms  ✓ LuxLib
   9040.5 ms  ✓ Lux
  94 dependencies successfully precompiled in 36 seconds. 11 already precompiled.

Now let us control the randomness in our code using proper Pseudo Random Number Generator (PRNG)

julia

rng = Random.default_rng()
Random.seed!(rng, 0)

Random.TaskLocalRNG()

Arrays

The starting point for all of our models is the Array (sometimes referred to as a Tensor in other frameworks). This is really just a list of numbers, which might be arranged into a shape like a square. Let's write down an array with three elements.

julia

x = [1, 2, 3]

3-element Vector{Int64}:
 1
 2
 3

Here's a matrix – a square array with four elements.

julia

x = [1 2; 3 4]

2×2 Matrix{Int64}:
 1  2
 3  4

We often work with arrays of thousands of elements, and don't usually write them down by hand. Here's how we can create an array of 5×3 = 15 elements, each a random number from zero to one.

julia

x = rand(rng, 5, 3)

5×3 Matrix{Float64}:
 0.455238   0.746943   0.193291
 0.547642   0.746801   0.116989
 0.773354   0.97667    0.899766
 0.940585   0.0869468  0.422918
 0.0296477  0.351491   0.707534

There's a few functions like this; try replacing rand with ones, zeros, or randn.

By default, Julia works stores numbers is a high-precision format called Float64. In ML we often don't need all those digits, and can ask Julia to work with Float32 instead. We can even ask for more digits using BigFloat.

julia

x = rand(BigFloat, 5, 3)

5×3 Matrix{BigFloat}:
 0.981339    0.793159  0.459019
 0.043883    0.624384  0.56055
 0.164786    0.524008  0.0355555
 0.414769    0.577181  0.621958
 0.00823197  0.30215   0.655881

julia

x = rand(Float32, 5, 3)

5×3 Matrix{Float32}:
 0.567794   0.369178   0.342539
 0.0985227  0.201145   0.587206
 0.776598   0.148248   0.0851708
 0.723731   0.0770206  0.839303
 0.404728   0.230954   0.679087

We can ask the array how many elements it has.

julia

length(x)

Or, more specifically, what size it has.

julia

size(x)

(5, 3)

We sometimes want to see some elements of the array on their own.

julia

5×3 Matrix{Float32}:
 0.567794   0.369178   0.342539
 0.0985227  0.201145   0.587206
 0.776598   0.148248   0.0851708
 0.723731   0.0770206  0.839303
 0.404728   0.230954   0.679087

julia

x[2, 3]

0.58720636f0

This means get the second row and the third column. We can also get every row of the third column.

julia

x[:, 3]

5-element Vector{Float32}:
 0.34253937
 0.58720636
 0.085170805
 0.8393034
 0.67908657

We can add arrays, and subtract them, which adds or subtracts each element of the array.

julia

x + x

5×3 Matrix{Float32}:
 1.13559   0.738356  0.685079
 0.197045  0.40229   1.17441
 1.5532    0.296496  0.170342
 1.44746   0.154041  1.67861
 0.809456  0.461908  1.35817

julia

x - x

5×3 Matrix{Float32}:
 0.0  0.0  0.0
 0.0  0.0  0.0
 0.0  0.0  0.0
 0.0  0.0  0.0
 0.0  0.0  0.0

Julia supports a feature called broadcasting, using the . syntax. This tiles small arrays (or single numbers) to fill bigger ones.

julia

x .+ 1

5×3 Matrix{Float32}:
 1.56779  1.36918  1.34254
 1.09852  1.20114  1.58721
 1.7766   1.14825  1.08517
 1.72373  1.07702  1.8393
 1.40473  1.23095  1.67909

We can see Julia tile the column vector 1:5 across all rows of the larger array.

julia

zeros(5, 5) .+ (1:5)

5×5 Matrix{Float64}:
 1.0  1.0  1.0  1.0  1.0
 2.0  2.0  2.0  2.0  2.0
 3.0  3.0  3.0  3.0  3.0
 4.0  4.0  4.0  4.0  4.0
 5.0  5.0  5.0  5.0  5.0

The x' syntax is used to transpose a column 1:5 into an equivalent row, and Julia will tile that across columns.

julia

zeros(5, 5) .+ (1:5)'

5×5 Matrix{Float64}:
 1.0  2.0  3.0  4.0  5.0
 1.0  2.0  3.0  4.0  5.0
 1.0  2.0  3.0  4.0  5.0
 1.0  2.0  3.0  4.0  5.0
 1.0  2.0  3.0  4.0  5.0

We can use this to make a times table.

julia

(1:5) .* (1:5)'

5×5 Matrix{Int64}:
 1   2   3   4   5
 2   4   6   8  10
 3   6   9  12  15
 4   8  12  16  20
 5  10  15  20  25

Finally, and importantly for machine learning, we can conveniently do things like matrix multiply.

julia

W = randn(5, 10)
x = rand(10)
W * x

5-element Vector{Float64}:
  1.2197981041108443
 -2.62625877100596
 -2.8573820474674845
 -2.4319346874291314
  1.0108668577150213

Julia's arrays are very powerful, and you can learn more about what they can do here.

CUDA Arrays

CUDA functionality is provided separately by the CUDA.jl package. If you have a GPU and LuxCUDA is installed, Lux will provide CUDA capabilities. For additional details on backends see the manual section.

You can manually add CUDA. Once CUDA is loaded you can move any array to the GPU with the cu function (or the gpu function exported by `Lux``), and it supports all of the above operations with the same syntax.

julia

using LuxCUDA

if LuxCUDA.functional()
    x_cu = cu(rand(5, 3))
    @show x_cu
end

(Im)mutability

Lux as you might have read is Immutable by convention which means that the core library is built without any form of mutation and all functions are pure. However, we don't enforce it in any form. We do strongly recommend that users extending this framework for their respective applications don't mutate their arrays.

julia

x = reshape(1:8, 2, 4)

2×4 reshape(::UnitRange{Int64}, 2, 4) with eltype Int64:
 1  3  5  7
 2  4  6  8

To update this array, we should first copy the array.

julia

x_copy = copy(x)
view(x_copy, :, 1) .= 0

println("Original Array ", x)
println("Mutated Array ", x_copy)

Original Array [1 3 5 7; 2 4 6 8]
Mutated Array [0 3 5 7; 0 4 6 8]

Note that our current default AD engine (Zygote) is unable to differentiate through this mutation, however, for these specialized cases it is quite trivial to write custom backward passes. (This problem will be fixed once we move towards Enzyme.jl)

Managing Randomness

We rely on the Julia StdLib Random for managing the randomness in our execution. First, we create an PRNG (pseudorandom number generator) and seed it.

julia

rng = Xoshiro(0)     # Creates a Xoshiro PRNG with seed 0

Random.Xoshiro(0xdb2fa90498613fdf, 0x48d73dc42d195740, 0x8c49bc52dc8a77ea, 0x1911b814c02405e8, 0x22a21880af5dc689)

If we call any function that relies on rng and uses it via randn, rand, etc. rng will be mutated. As we have already established we care a lot about immutability, hence we should use Lux.replicate on PRNGs before using them.

First, let us run a random number generator 3 times with the replicated rng.

julia

random_vectors = Vector{Vector{Float64}}(undef, 3)
for i in 1:3
    random_vectors[i] = rand(Lux.replicate(rng), 10)
    println("Iteration $i ", random_vectors[i])
end
@assert random_vectors[1] ≈ random_vectors[2] ≈ random_vectors[3]

Iteration 1 [0.4552384158732863, 0.5476424498276177, 0.7733535276924052, 0.9405848223512736, 0.02964765308691042, 0.74694291453392, 0.7468008914093891, 0.9766699015845924, 0.08694684883050086, 0.35149138733595564]
Iteration 2 [0.4552384158732863, 0.5476424498276177, 0.7733535276924052, 0.9405848223512736, 0.02964765308691042, 0.74694291453392, 0.7468008914093891, 0.9766699015845924, 0.08694684883050086, 0.35149138733595564]
Iteration 3 [0.4552384158732863, 0.5476424498276177, 0.7733535276924052, 0.9405848223512736, 0.02964765308691042, 0.74694291453392, 0.7468008914093891, 0.9766699015845924, 0.08694684883050086, 0.35149138733595564]

As expected we get the same output. We can remove the replicate call and we will get different outputs.

julia

for i in 1:3
    println("Iteration $i ", rand(rng, 10))
end

Iteration 1 [0.4552384158732863, 0.5476424498276177, 0.7733535276924052, 0.9405848223512736, 0.02964765308691042, 0.74694291453392, 0.7468008914093891, 0.9766699015845924, 0.08694684883050086, 0.35149138733595564]
Iteration 2 [0.018743665453639813, 0.8601828553599953, 0.6556360448565952, 0.7746656838366666, 0.7817315740767116, 0.5553797706980106, 0.1261990389976131, 0.4488101521328277, 0.624383955429775, 0.05657739601024536]
Iteration 3 [0.19597391412112541, 0.6830945313415872, 0.6776220912718907, 0.6456416023530093, 0.6340362477836592, 0.5595843665394066, 0.5675557670686644, 0.34351700231383653, 0.7237308297251812, 0.3691778381831775]

Automatic Differentiation

Julia has quite a few (maybe too many) AD tools. For the purpose of this tutorial, we will use:

ForwardDiff.jl – For Jacobian-Vector Product (JVP)
Zygote.jl – For Vector-Jacobian Product (VJP)

Slight Detour: We have had several questions regarding if we will be considering any other AD system for the reverse-diff backend. For now we will stick to Zygote.jl, however once we have tested Lux extensively with Enzyme.jl, we will make the switch.

Even though, theoretically, a VJP (Vector-Jacobian product - reverse autodiff) and a JVP (Jacobian-Vector product - forward-mode autodiff) are similar—they compute a product of a Jacobian and a vector—they differ by the computational complexity of the operation. In short, when you have a large number of parameters (hence a wide matrix), a JVP is less efficient computationally than a VJP, and, conversely, a JVP is more efficient when the Jacobian matrix is a tall matrix.

julia

using ComponentArrays, ForwardDiff, Zygote

Precompiling ComponentArrays...
    842.1 ms  ✓ ComponentArrays
  1 dependency successfully precompiled in 1 seconds. 23 already precompiled.
Precompiling MLDataDevicesComponentArraysExt...
    457.2 ms  ✓ MLDataDevices → MLDataDevicesComponentArraysExt
  1 dependency successfully precompiled in 0 seconds. 26 already precompiled.
Precompiling LuxComponentArraysExt...
    467.9 ms  ✓ ComponentArrays → ComponentArraysOptimisersExt
   1355.5 ms  ✓ Lux → LuxComponentArraysExt
   1960.5 ms  ✓ ComponentArrays → ComponentArraysKernelAbstractionsExt
  3 dependencies successfully precompiled in 2 seconds. 107 already precompiled.
Precompiling Zygote...
    350.7 ms  ✓ IteratorInterfaceExtensions
    466.1 ms  ✓ DataValueInterfaces
    494.8 ms  ✓ RealDot
    507.4 ms  ✓ DataAPI
    610.9 ms  ✓ OrderedCollections
    627.0 ms  ✓ Serialization
    630.2 ms  ✓ SuiteSparse_jll
    677.9 ms  ✓ AbstractFFTs
    377.7 ms  ✓ TableTraits
   1064.2 ms  ✓ FillArrays
   1048.9 ms  ✓ ZygoteRules
    439.2 ms  ✓ AbstractFFTs → AbstractFFTsChainRulesCoreExt
    393.9 ms  ✓ FillArrays → FillArraysStatisticsExt
    793.7 ms  ✓ Tables
   1947.9 ms  ✓ IRTools
    698.6 ms  ✓ StructArrays
   1725.8 ms  ✓ Distributed
    398.5 ms  ✓ StructArrays → StructArraysAdaptExt
    417.3 ms  ✓ StructArrays → StructArraysLinearAlgebraExt
   3702.3 ms  ✓ SparseArrays
    613.8 ms  ✓ SuiteSparse
    637.6 ms  ✓ Statistics → SparseArraysExt
    636.2 ms  ✓ Adapt → AdaptSparseArraysExt
    651.3 ms  ✓ ChainRulesCore → ChainRulesCoreSparseArraysExt
    655.4 ms  ✓ StructArrays → StructArraysSparseArraysExt
    694.2 ms  ✓ FillArrays → FillArraysSparseArraysExt
    586.8 ms  ✓ SparseInverseSubset
   5232.4 ms  ✓ ChainRules
  25261.6 ms  ✓ Zygote
  29 dependencies successfully precompiled in 36 seconds. 36 already precompiled.
Precompiling ArrayInterfaceSparseArraysExt...
    588.8 ms  ✓ ArrayInterface → ArrayInterfaceSparseArraysExt
  1 dependency successfully precompiled in 1 seconds. 8 already precompiled.
Precompiling SparseArraysExt...
    908.0 ms  ✓ KernelAbstractions → SparseArraysExt
  1 dependency successfully precompiled in 1 seconds. 27 already precompiled.
Precompiling MLDataDevicesSparseArraysExt...
    631.1 ms  ✓ MLDataDevices → MLDataDevicesSparseArraysExt
  1 dependency successfully precompiled in 1 seconds. 18 already precompiled.
Precompiling StructArraysStaticArraysExt...
    628.0 ms  ✓ StructArrays → StructArraysStaticArraysExt
  1 dependency successfully precompiled in 1 seconds. 19 already precompiled.
Precompiling StructArraysGPUArraysCoreExt...
    678.9 ms  ✓ StructArrays → StructArraysGPUArraysCoreExt
  1 dependency successfully precompiled in 1 seconds. 34 already precompiled.
Precompiling ArrayInterfaceChainRulesExt...
    737.9 ms  ✓ ArrayInterface → ArrayInterfaceChainRulesExt
  1 dependency successfully precompiled in 1 seconds. 40 already precompiled.
Precompiling MLDataDevicesChainRulesExt...
    771.6 ms  ✓ MLDataDevices → MLDataDevicesChainRulesExt
  1 dependency successfully precompiled in 1 seconds. 41 already precompiled.
Precompiling MLDataDevicesFillArraysExt...
    415.6 ms  ✓ MLDataDevices → MLDataDevicesFillArraysExt
  1 dependency successfully precompiled in 0 seconds. 15 already precompiled.
Precompiling MLDataDevicesZygoteExt...
   1465.3 ms  ✓ MLDataDevices → MLDataDevicesZygoteExt
  1 dependency successfully precompiled in 2 seconds. 71 already precompiled.
Precompiling LuxZygoteExt...
   2454.9 ms  ✓ Lux → LuxZygoteExt
  1 dependency successfully precompiled in 3 seconds. 143 already precompiled.
Precompiling ComponentArraysZygoteExt...
   1445.9 ms  ✓ ComponentArrays → ComponentArraysZygoteExt
  1 dependency successfully precompiled in 2 seconds. 77 already precompiled.

Gradients

For our first example, consider a simple function computing $f (x) = \frac{1}{2} x^{T} x$ , where $\nabla f (x) = x$

julia

f(x) = x' * x / 2
∇f(x) = x  # `∇` can be typed as `\nabla<TAB>`
v = randn(rng, Float32, 4)

4-element Vector{Float32}:
 -0.4051151
 -0.4593922
  0.92155594
  1.1871622

Let's use ForwardDiff and Zygote to compute the gradients.

julia

println("Actual Gradient: ", ∇f(v))
println("Computed Gradient via Reverse Mode AD (Zygote): ", only(Zygote.gradient(f, v)))
println("Computed Gradient via Forward Mode AD (ForwardDiff): ", ForwardDiff.gradient(f, v))

Actual Gradient: Float32[-0.4051151, -0.4593922, 0.92155594, 1.1871622]
Computed Gradient via Reverse Mode AD (Zygote): Float32[-0.4051151, -0.4593922, 0.92155594, 1.1871622]
Computed Gradient via Forward Mode AD (ForwardDiff): Float32[-0.4051151, -0.4593922, 0.92155594, 1.1871622]

Note that AD.gradient will only work for scalar valued outputs.

Jacobian-Vector Product

I will defer the discussion on forward-mode AD to https://book.sciml.ai/notes/08-Forward-Mode_Automatic_Differentiation_(AD)_via_High_Dimensional_Algebras/. Here let us just look at a mini example on how to use it.

julia

f(x) = x .* x ./ 2
x = randn(rng, Float32, 5)
v = ones(Float32, 5)

5-element Vector{Float32}:
 1.0
 1.0
 1.0
 1.0
 1.0

Using DifferentiationInterface

While DifferentiationInterface provides these functions for a wider range of backends, we currently don't recommend using them with Lux models, since the functions presented here come with additional goodies like fast second-order derivatives.

Compute the jvp. AutoForwardDiff specifies that we want to use ForwardDiff.jl for the Jacobian-Vector Product

julia

jvp = jacobian_vector_product(f, AutoForwardDiff(), x, v)
println("JVP: ", jvp)

JVP: Float32[-0.877497, 1.1953009, -0.057005208, 0.25055695, 0.09351656]

Vector-Jacobian Product

Using the same function and inputs, let us compute the VJP.

julia

vjp = vector_jacobian_product(f, AutoZygote(), x, v)
println("VJP: ", vjp)

VJP: Float32[-0.877497, 1.1953009, -0.057005208, 0.25055695, 0.09351656]

Linear Regression

Finally, now let us consider a linear regression problem. From a set of data-points ${(x_{i}, y_{i}), i \in {1, \dots, k}, x_{i} \in R^{n}, y_{i} \in R^{m}}$ , we try to find a set of parameters $W$ and $b$ , s.t. $f_{W, b} (x) = W x + b$ , which minimizes the mean squared error:

L (W, b) ⟶ \sum_{i = 1}^{k} \frac{1}{2} ∥ y_{i} - f_{W, b} (x_{i}) ∥_{2}^{2}

We can write f from scratch, but to demonstrate Lux, let us use the Dense layer.

julia

model = Dense(10 => 5)

rng = Random.default_rng()
Random.seed!(rng, 0)

Random.TaskLocalRNG()

Let us initialize the parameters and states (in this case it is empty) for the model.

julia

ps, st = Lux.setup(rng, model)
ps = ComponentArray(ps)

ComponentVector{Float32}(weight = Float32[-0.48351598 0.29944375 0.44048917 0.5221656 0.20001543 0.1437841 4.8317274f-6 0.5310851 -0.30674052 0.034259234; -0.04903387 -0.4242767 0.27051234 0.40789893 -0.43846482 -0.17706361 -0.03258145 0.46514034 0.1958431 0.23992883; 0.45016125 0.48263642 -0.2990853 -0.18695377 -0.11023762 -0.4418456 0.40354207 0.25278285 0.18056087 -0.3523193; 0.05218964 -0.09701932 0.27035674 0.12589 -0.29561827 0.34717593 -0.42189494 -0.13073668 0.36829436 -0.3097294; 0.20277858 -0.51524514 -0.22635892 0.18841726 0.29828635 0.21690917 -0.04265762 -0.41919118 0.071482725 -0.45247704], bias = Float32[-0.04199602, -0.093925126, -0.0007736237, -0.19397983, 0.0066712513])

Set problem dimensions.

julia

n_samples = 20
x_dim = 10
y_dim = 5

Generate random ground truth W and b.

julia

W = randn(rng, Float32, y_dim, x_dim)
b = randn(rng, Float32, y_dim)

5-element Vector{Float32}:
 -0.9436797
  1.5164032
  0.011937321
  1.4339262
 -0.2771789

Generate samples with additional noise.

julia

x_samples = randn(rng, Float32, x_dim, n_samples)
y_samples = W * x_samples .+ b .+ 0.01f0 .* randn(rng, Float32, y_dim, n_samples)
println("x shape: ", size(x_samples), "; y shape: ", size(y_samples))

x shape: (10, 20); y shape: (5, 20)

For updating our parameters let's use Optimisers.jl. We will use Stochastic Gradient Descent (SGD) with a learning rate of 0.01.

julia

using Optimisers, Printf

Define the loss function

julia

lossfn = MSELoss()

println("Loss Value with ground true parameters: ", lossfn(W * x_samples .+ b, y_samples))

Loss Value with ground true parameters: 9.3742405e-5

We will train the model using our training API.

julia

function train_model!(model, ps, st, opt, nepochs::Int)
    tstate = Training.TrainState(model, ps, st, opt)
    for i in 1:nepochs
        grads, loss, _, tstate = Training.single_train_step!(
            AutoZygote(), lossfn, (x_samples, y_samples), tstate
        )
        if i % 1000 == 1 || i == nepochs
            @printf "Loss Value after %6d iterations: %.8f\n" i loss
        end
    end
    return tstate.model, tstate.parameters, tstate.states
end

model, ps, st = train_model!(model, ps, st, Descent(0.01f0), 10000)

println("Loss Value after training: ", lossfn(first(model(x_samples, ps, st)), y_samples))

Loss Value after      1 iterations: 7.80465555
Loss Value after   1001 iterations: 0.12477568
Loss Value after   2001 iterations: 0.02535537
Loss Value after   3001 iterations: 0.00914141
Loss Value after   4001 iterations: 0.00407581
Loss Value after   5001 iterations: 0.00198415
Loss Value after   6001 iterations: 0.00101147
Loss Value after   7001 iterations: 0.00053332
Loss Value after   8001 iterations: 0.00029203
Loss Value after   9001 iterations: 0.00016878
Loss Value after  10000 iterations: 0.00010551
Loss Value after training: 0.00010546855

Appendix

julia

using InteractiveUtils
InteractiveUtils.versioninfo()

if @isdefined(MLDataDevices)
    if @isdefined(CUDA) && MLDataDevices.functional(CUDADevice)
        println()
        CUDA.versioninfo()
    end

    if @isdefined(AMDGPU) && MLDataDevices.functional(AMDGPUDevice)
        println()
        AMDGPU.versioninfo()
    end
end

Julia Version 1.11.5
Commit 760b2e5b739 (2025-04-14 06:53 UTC)
Build Info:
  Official https://julialang.org/ release
Platform Info:
  OS: Linux (x86_64-linux-gnu)
  CPU: 128 × AMD EPYC 7502 32-Core Processor
  WORD_SIZE: 64
  LLVM: libLLVM-16.0.6 (ORCJIT, znver2)
Threads: 16 default, 0 interactive, 8 GC (on 16 virtual cores)
Environment:
  JULIA_CPU_THREADS = 16
  JULIA_DEPOT_PATH = /cache/julia-buildkite-plugin/depots/01872db4-8c79-43af-ab7d-12abac4f24f6
  JULIA_PKG_SERVER = 
  JULIA_NUM_THREADS = 16
  JULIA_CUDA_HARD_MEMORY_LIMIT = 100%
  JULIA_PKG_PRECOMPILE_AUTO = 0
  JULIA_DEBUG = Literate

This page was generated using Literate.jl.

Trusted by

Julia & Lux for the Uninitiated ​

Arrays ​

CUDA Arrays ​

(Im)mutability ​

Managing Randomness ​

Automatic Differentiation ​

Gradients ​

Jacobian-Vector Product ​

Vector-Jacobian Product ​

Linear Regression ​

Appendix ​

Julia & Lux for the Uninitiated

Arrays

CUDA Arrays

(Im)mutability

Managing Randomness

Automatic Differentiation

Gradients

Jacobian-Vector Product

Vector-Jacobian Product

Linear Regression

Appendix