Description
Lots of algorithms that rely on randomization offer an argument random_state through which it's possible to pass them RNG seeds that would ensure reproducible results. The IterativeImputer class will take some regressor/classifier and set their attribute random_state to a NumPy MT19937 object class if the object has such attribute, in such a way that this object is modified implicitly once the regressor/classifier to which it was assigned produces some random number.
This is fine for SciKit-Learn’s own classes, but other libraries which provide different regressors/classifiers which are SciKit-Learn-compatible (at least according to tests) might not be able to use such RandomState objects (e.g. if they generate random numbers in C++ with some method other than MT19937).
For example, XGBoost and LightGBM SciKit-Learn-compatible classes will only accept integers as parameters for their random_state, and there’s potentially many others too which will only accept integers.
Would be nice if IterativeImputer would instead generate a random integer seed to pass to the regressor/classifier at each iteration, so that it could be used with more external libraries.