hyperband_n_configs

(<code>numeric(1)</code>)
Lower bound of budget parameter.

r_min

(<code>numeric(1)</code>)
Upper bound of budget parameter.

r_max

(<code>numeric(1)</code>)
Fraction parameter of the successive halving algorithm: With every stage the
configuration budget is increased by a factor of <code>eta</code> and only the best
<code>1/eta</code> points are used for the next stage. Non-integer values are supported,
but <code>eta</code> is not allowed to be less or equal 1.

Calculates how many different configurations are sampled.

Implements hyperband method for hyperparameter
tuning. Various termination criteria can be set and combined. The
class 'AutoTuner' provides a convenient way to perform nested
resampling in combination with 'mlr3'. The hyperband algorithm was
proposed by Lisha Li, Kevin Jamieson, Giulia DeSalvo, Afshin
Rostamizadeh and Ameet Talwalkar (2018) <arXiv:1603.06560>.

Marc Becker

hyperband_n_configs: Hyperband Configs

Description

Usage

Arguments

Value