GH version of load_results has different behaviour than pip-installable version

[Kamuish]

When running the

 res = kima.load_results(diagnostic=False)

we get the traceback

Traceback (most recent call last):
  File "/home/amiguel/first_author_articles/WG1targets_Andre/kima_analysis/plot_kima.py", line 23, in <module>
    res = kima.load_results(diagnostic=False)
TypeError: load_results() missing 1 required positional argument: 'model_or_file'

An attempt to pass the current directory through the model_or_file argument leads to:

IsADirectoryError: [Errno 21] Is a directory: '.'

The same issue occurs when directly calling KimaResults that -- from the docs -- should be able to load data from the current directory

res = KimaResults.load(diagnostic=False)

PS: all of this was run with the context manager from kima.pykima.utils.chdir

[j-faria]

Currently (on main) you must call load_results as

res = kima.load_results(model, diagnostic=False)

but I will keep this issue open because I want to restore the ability to "guess" the model from the directory where the function runs.

[Kamuish]

Do you have to re-define all of the priors (and known_planets) or can kima infer those from the disk products?

[j-faria]

Those settings are stored in (both) the model and the kima_model_setup.txt file, so they should be inferred.

[Kamuish]

It seems that that is the case. For the loading, I guess that it is fairly straightforward. I am running it like this:

model_map = {"GPmodel": GPmodel, "RVFWHMmodel": RVFWHMmodel}
selected_folder = paths["run4"]
with chdir(selected_folder):
    with open(selected_folder / "kima_model_setup.txt") as to_read:
        for line in to_read:
            if "model" in line:
                Kimamodel = model_map[line.split(":")[-1].strip()]
            if "fix" in line:
                state = line.split(":")[-1].strip()
                fix = False if state == "false" else True
            if "npmax" in line:
                npmax = int(line.split(":")[-1])
            if "files" in line:
                input_files = line.split(":")[-1].split(",")[:-1]
                break

    data = RVData(
        filenames=input_files,
        units="ms",
        skip=0,
        indicators=["fwhm", "s-fwhm", "cont", "s-cont", "bis", "s-bis"],
    )

    model = Kimamodel(fix=fix, npmax=npmax, data=data)

    res = kima.load_results(model, diagnostic=False)

I will not open a PR due to:

1. I am fairly certain that such an approach is missing a few edge cases
2. not exactly sure if this functionality should be placed directly in the KimaResults objects or in the load_results function