JuliaAI / MLJTuning.jl

Hyperparameter optimization algorithms for use in the MLJ machine learning framework
MIT License
66 stars 12 forks source link

Make a tuned model's number of iterations, n, available to setup method #106

Closed ablaom closed 3 years ago

ablaom commented 3 years ago

Latin hypercube strategy needs to know the total iteration count up-front but it cannot get this information. The temporary solution in #96 is to introduce a separate n_max parameter for the Latin strategy, but there is little use in this having a different value.

So the proposal is to add n to the signature of setup:

state = setup(tuning::MyTuningStrategy, model, range, n, verbosity)