automl / neps

Neural Pipeline Search (NePS): Helps deep learning experts find the best neural pipeline.
https://automl.github.io/neps/
Apache License 2.0
44 stars 11 forks source link

Enhancing Flexibility in Priorband Usage: Allowing Partial Knowledge Without Mandatory Default Parameters #28

Open danrgll opened 8 months ago

danrgll commented 8 months ago

It would be beneficial to have the ability to add partial knowledge to the configuration area, especially for using Priorband, without having to set the "default" argument for all the parameters.

Example:

pipeline_space = dict(
    hyperparameter_a=neps.FloatParameter(lower=1e-5, upper=1e-1, log=True),
    hyperparameter_b=neps.IntegerParameter(lower=1, upper=20, is_fidelity=True), 
    hyperparameter_c=neps.IntegerParameter(lower=32, upper=128, **default=64, default_confidence="medium"**), # my knowledge 
)