GDAS Profile

class confopt.profile.GDASProfile(searchspace_type: str | SearchSpaceType, epochs: int, tau_min: float = 0.1, tau_max: float = 10, **kwargs: Any)

Bases: BaseProfile, ABC

Train Profile class for GDAS (Gradient-based search using Differentiable Architecture Sampler) It inherits from the BaseProfile class and sets the sampler config for DARTS sampler.

Variables:

  • tau_min (float) – Minimum temperature for sampling.

  • tau_max (float) – Maximum temperature for sampling.

configure_sampler(**kwargs) → None

Configures the sampler settings based on provided keyword arguments.

The configuration options are passed as keyword arguments.

Parameters:

**kwargs –

Arbitrary keyword arguments. Possible keys include:

sample_frequency (str): The rate at which samples should be taken

arch_combine_fn (str): The function used to combine architectures (str) when FairDARTS is used, this should be ‘sigmoid’. Default value is ‘default’.

tau_min (float): Minimum temperature for sampling.

tau_max (float): Maximum temperature for sampling.

Example

>>> from confopt.profile import GDASProfile
>>> gdas_profile = GDASProfile(searchspace='darts', epochs=50)
>>> gdas_profile.configure_sampler(tau_min=02, tau_max=20)

The accepted keyword arguments should align with the sampler’s configuration and the attributes can be configured dynamically.

Raises:

ValueError – If an unrecognized keyword is passed in kwargs.

Returns:

None