Define random sampling over a hyperparameter search space
In random sampling, hyperparameter values are randomly selected from the defined search space. Random sampling allows the search space to include both discrete and continuous hyperparameters.
random_parameter_sampling(parameter_space, properties = NULL)
parameter_space |
A named list containing each parameter and its
distribution, e.g. |
properties |
A named list of additional properties for the algorithm. |
The RandomParameterSampling object.
In this sampling algorithm, parameter values are chosen from a set of
discrete values or a distribution over a continuous range. Functions you can
use include:
choice(), randint(), uniform(), quniform(), loguniform(),
qloguniform(), normal(), qnormal(), lognormal(), and qlognormal().
choice(), randint(), uniform(), quniform(), loguniform(),
qloguniform(), normal(), qnormal(), lognormal(), qlognormal()
## Not run:
param_sampling <- random_parameter_sampling(list("learning_rate" = normal(10, 3),
"keep_probability" = uniform(0.05, 0.1),
"batch_size" = choice(c(16, 32, 64, 128))))
## End(Not run)Please choose more modern alternatives, such as Google Chrome or Mozilla Firefox.