scala - How to determine parameter ranges of MLlib algorithms? -
i'm developing spark project in scala. want create mllib models linear regression, naive bayes, gradient boosting regression etc.
here want try generating these models different combinations using parameters. example gradient boosting regression takes num_iters, maxdepth, learning_rate , maxbins parameter. specified parameter ranges algorithm:
num_iters -> list.range(3,10) maxdepth -> list.range(5,20) maxbins -> list.range(25,50) learning_rate -> list(0.000001, 0.00001, 0.0001, 0.001, 0.01, 0.1)
i searched lots of sources didn't find solution.
what want know how can specify these parameter's ranges logically.
Comments
Post a Comment