Array-like arguments in odr_fit cause an error instead of being converted to arrays

[nicolas-joffre]

If beta0 is a list instead of an array, I get an error message AttributeError: 'list' object has no attribute 'ndim'

If weight_x is a Pandas series instead of an array, I get an error message TypeError:`weight_x` must be a float or an array.

Both of these are array-like and should be converted to an array beforehand.

[HugoMVale]

Thanks for the feedback.
The type hints are very clear regrading the valid input types.
I kind of understand your point, but I am not currently convinced that converting inputs is always thing to do.
I will think about it, but do not promise anything.