please solve in Julia(in VS Code)

function huber_loss(y_pred::T, y_true::T; ::Float64 = 0.1) where T <: Real

return nothing

end

function huber_loss(y_pred::AbstractArray{T}, y_true::AbstractArray{T}) where T <: Real

return nothing

end

function unit_test_huber_loss()::Bool

Random.seed!(0)

@assert huber_loss(1.0,1.0) == 0.0

@assert huber_loss(1.0,2.0; = 0.9) == 0.49500000000000005

@assert isapprox(huber_loss(randn(100,100),randn(100,100)), 0.10791842, atol = 1e-2)

@assert isapprox(huber_loss(randn(100),randn(100)), 0.107945, atol = 1e-2)

@info "You can not stop now comrade!!! jump to the next exercise!!!"

return 1

end

## See you have implemented huber_loss() well??

unit_test_huber_loss()

### create a roof for the logistic regression LogisticClassifier

abstract type LinearRegression end

mutable struct linear_regression <: LinearRegression

## This part is given to you!!!

## Realize that we have fields: and bias.

::AbstractVector

bias::Real

linear_regression(n::Int64) = new(0.004*randn(n), zero(1))

end

### write the forwardpass function

function (lr::linear_regression)(X::Matrix{T}) where T<:Real

## This dude is the forward pass function!!!

return nothing

end

function unit_test_forward_pass()::Bool

try

linear_regression(20)(randn(10,20))

catch ERROR

error("SEG-FAULT!!!!")

end

@info "Real test started!!!!"

for i in ProgressBar(1:10000)

sleep(0.0001)

lr = linear_regression(3)

x = randn(2,3)

@assert lr(x) == x*lr. .+ lr.bias

end

@info "Oki doki!!!"

return 1

end

### Let's give a try!!!!!!

unit_test_forward_pass()

## we shall now implement fit! method!!!

## before we get ready run the next 5 lines to see in this setting grad function returns a dictionary:

#=

lr = linear_regression(10)

val, grad = Zygote.withgradient(lr) do lr

norm(lr.)^2 + lr.bias^2

end

=#

function update_grads!(lr::LinearRegression, learning_rate::Float64, grads)

## Here you will implement update_grads, this function returns nothing.

## Search for setfield! and getfield functions.

## x -= learning_rate*grads will happen here!!!

return nothing

end

function fit!(lr::linear_regression,

X::AbstractMatrix,

y::AbstractVector;

learning_rate::Float64 = 0.00001,

max_iter::Integer = 5,

::Float64 = 0.01)

##

return nothing

end

## Let's give a try!!!

lr = linear_regression(20)

X = randn(100, 20)

y = randn(100)

fit!(lr, X, y; learning_rate = 0.00001, max_iter = 10000)

### Things seem to work fine!!!

function unit_test_for_fit()

Random.seed!(0)

lr = linear_regression(20)

X = randn(100, 20)

y = randn(100)

fit!(lr, X, y; learning_rate = 0.0001, max_iter = 10000, = 0.1)

@assert norm(lr.)^2 + lr.bias^2 < 0.01 "Your penalty method does not work!!!"

@assert mean((lr(X) - y).^2) < 1.3 "Yo do not fit perfectly!!!!"

@info "Okito dokito buddy!!!"

return 1

end

## Run next line to see what happens??? ##

unit_test_for_fit()

## -- ##