Get the number of iterations needed to train a model

[Genarito]

Is your feature request related to a problem? Please describe.
I am doing some training time measurements of the FastKernelSurvivalSVM model. The problem is that I would like to know how much time it takes for each iteration, but I don't know this amount. It would be nice if the model has the attribute to be able to access it.

Describe the solution you'd like
Expose a new attribute (for example, n_iter, similarly to modern Scikit-Learn versions, where they added the n_iter_ attribute).

Describe alternatives you've considered
I'm accessing the number of iterations through the optimizer_result_ attribute:

estimator = FastKernelSurvivalSVM(rank_ratio=0.0, max_iter=1000, tol=1e-5)
# Trains the model...
estimator.optimizer_result_.nit

But I'm not really convinced if what I'm doing is right. It'd be cool to have an official solution.

Thanks in advance, this is a great lib ❤️

[sebp]

You inferred it correctly, estimator.optimizer_result_.nit is what you want (see the scipy docs for details).

I agree that having an n_iter_ attribute would be useful to match sklearn.

[Genarito]

Cool! Thank you very much!