hype {comparer}

Hyperparameter optimization

Description

Hyperparameter optimization

Usage

hype(
  eval_func,
  ...,
  X0 = NULL,
  Z0 = NULL,
  n_lhs,
  extract_output_func,
  verbose = 1,
  model = "GauPro",
  covtype = "matern5_2",
  nugget.estim = TRUE
)

Arguments

eval_func

The function we evaluate.

...

Pass in hyperparameters, such as par_unif() as unnamed arguments.

X0

A data frame of initial points to include. They must have the same names as the hyperparameters. If Z0 is also passed, it should match the points in X0. If Z0 is not passed, then X0 will be the first points evaluated.

Z0

A vector whose values are the result of applying 'eval_func' to each row of X0.

n_lhs

The number of random points to start with. They are selected using a Latin hypercube sample.

extract_output_func

A function that takes in the output from 'eval_func' and returns the value we are trying to minimize.

verbose

How much should be printed? 0 is none, 1 is standard, 2 is more, 5+ is a lot

model

What kind of model to use.

covtype

The covariance function to use for the Gaussian process model.

nugget.estim

Whether a nugget should be estimated when fitting the Gaussian process model.

Examples

# Have df output, but only use one value from it
h1 <- hype(
  eval_func = function(a, b) {data.frame(c=a^2+b^2, d=1:2)},
  extract_output_func = function(odf) {odf$c[1]},
  a = par_unif('a', -1, 2),
  b = par_unif('b', -10, 10),
  n_lhs = 10
)
h1$run_all()
h1$add_EI(n = 1)
h1$run_all()
#system.time(h1$run_EI_for_time(sec=3, batch_size = 1))
#system.time(h1$run_EI_for_time(sec=3, batch_size = 3))
h1$plotorder()
h1$plotX()