А, ну значит всё правильно.

Недавно видел пост про то что Arrow теперь в Джулии поддерживается, с новой версии, что он бинарный, колончатый и с метаданными, но юзкейс до конца не понял - он дисковый для чтения в память или в нем самом как-то вычисления организуются более быстро

Ну и выдержка из первого блога


The Arrow format was created in 2016 by Wes McKinney, the creator of the extremely popular pandas Python package, to produce language-independent open standards and libraries to accelerate and simplify in-memory computing. [6] Of note, it is ideal for columnar data, cache-efficiency on CPUs and GPUs, leverages parallel processing, and optimized for scan and random access. Arrow.jl brilliantly used memory mapping to implement the standard in an immensely efficient way.

It is a name of variable in my model

I enter Xd and Yd  by using following codes

It looks like Data[blablabla] creates two dimensional array of the type Any. You should extract column that you wish to draw and convert it something like Vector{Float64}.

It is impossible to say exactly what type you should use, since only you know what kind of data you are using.

See this

And it is my excel file:

It should be typeof(Xd), not typeof("Xd")

Oh yes
I do it

Yes, you should extract necessary column and convert it to vector of floats

How?

Do you know how to manipulate arrays?

How to make slices?