I assume you know which distributions are possible once you are inside your function? I would probably just pass a parameter for the distribution name and then a set of parameters in a Python dictionary.

This method also allows for pretty easy testing, because you know that distributions expects only certain parameters, so you can also add small checks to confirm everything is available.

def my_distribution(dist_name, dist_params):
    if dist_name == "normal":
        assert ("mean" in params.keys()) and ("variance" in params.keys()), "Missing expected parameters for {} distribution".format(dist_name)
    # Perform some checks for other distributions as necessary...

    # perform your own steps...

Now you use it like this:

# assume these are provided by your earlier code
distr = "normal"
params = {"mean": 5.0, "variance": 2.0}

result = my_distribution(dist_name=distr, dist_params=params)

You could of course make classes that hold all configurations and checks, but I would argue you are starting to make things unnecessarily complicated, as Numpy does so much for you already with it's built-in distribution capabilities.

I assume you know which distributions are possible once you are inside your function? I would probably just pass a parameter for the distribution name and then a set of parameters in a Python dictionary.

This method also allows for pretty easy testing, because you know that distributions expects only certain parameters, so you can also add small checks to confirm everything is available.

def my_distribution(dist_name, dist_params):
    if dist_name == "normal":
        assert ("mean" in params.keys()) and ("variance" in params.keys()), "Missing expected parameters for {} distribution".format(dist_name)
    # Perform some checks for other distributions as necessary...

    # perform your own steps...

Now you use it like this:

# assume these are provided by your earlier code
distr = "normal"
params = {"mean": 5.0, "variance": 2.0}

result = my_distribution(dist_name=distr, dist_params=params)

Edit:

Here is an example of a single function that can handle multiple distributions in a single call:

def my_distributions(dist_collection):
    # Perform some checks for other distributions as necessary...
    allowed_dists = ["normal", "uniform", "dirichlet", "rayleigh"]
    assert all(
        dist_name in allowed_dists for dist_name in dist_collection
    ), "Input contains disallowed distribution"

    # Do something for each distribution name with its parameters:
    for dist_name, dist_params in params.items():
        print(dist_name, dist_params)

Now the input needs to be specified a little differently:

all_distributions = {
    "normal": {"loc": 1, "scale": 2, "size": 100},
    "uniform": {"low": 0, "high": 1, "size": 100},
}

A single call can now work over many distributions:

my_distributions(all_distributions)

You could of course make classes that hold all configurations and checks, but I would argue you are starting to make things unnecessarily complicated, as Numpy does so much for you already with it's built-in distribution capabilities.