What is L in the custom objectives?

[unary-code]

Where does the "L" in L(Inf) come from?
How can I return a "L(Inf)" value in my Julia code in my julia helper object jl?
Currently, even when I do "using SymbolicRegression" in my Julia helper object, I can not return "L(Inf)" in my custom function "eval_loss", because when I do return "L(Inf)" inside "eval_loss" and I run the julia code of the custom function "eval_loss", the Julia interpreter gives me a error in the terminal.

For your reference, "jl" is my julia helper object, and "jl" is defined as

jl = pysr.julia_helpers.init_julia(julia_kwargs={"threads": 8, "optimize": 3})

Specifically, when I run:

jl.eval("""
using SymbolicRegression
function my_func(some_x)

    return L(Inf)
end

println(my_func(3))

""")

in my Python3 terminal, the Julia interpreter gives me this error in the terminal:

Traceback (most recent call last):
  File "<stdin>", line 1, in <module>
  File "/opt/anaconda3/envs/pysr_env/lib/python3.10/site-packages/julia/core.py", line 627, in eval
    ans = self._call(src)
  File "/opt/anaconda3/envs/pysr_env/lib/python3.10/site-packages/julia/core.py", line 555, in _call
    self.check_exception(src)
  File "/opt/anaconda3/envs/pysr_env/lib/python3.10/site-packages/julia/core.py", line 609, in check_exception
    raise JuliaError(u'Exception \'{}\' occurred while calling julia code:\n{}'
julia.core.JuliaError: Exception 'UndefVarError: `L` not defined' occurred while calling julia code:

using SymbolicRegression
function my_func(some_x)

    return L(Inf)
end

println(my_func(3))

Why this is somewhat important: When I pass into my Julia objective function a lot of the math functions that my model.equations_ contains, the Julia helper's objective function returns a different loss value from what the model.equations_ reports as the loss value in the "loss" column for that math function.
So, I want to make my Julia helper's objective function (that I use to double-check the model's reported value for loss for a particular math function) as close as possible to the Julia objective function that my model "model" uses.
To make my Julia helper's objective function as close as possible to my model's objective function, I want to be able (in my Julia helper's objective function) to return L(Inf) instead of just Inf.

[MilesCranmer]

L is just a type variable that you have in the loss functions — it’s based on the dataset type of Dataset{T,L} that you would need to declare in the function signature. L could be a Float32 or a Float64 or something else.

Note that L(Inf) just means: “convert infinity to type L”.

If you want to use L from the loss function in other functions, you could pass it. But normally all you need to do is just wrap the output of the other function with L(…) to convert whatever numerical type it outputs.

[MilesCranmer]

If you are using precision=32, you could also just assume L=Float32 and use that instead