DistributedHyperOpt.jl

DistributedHyperOpt.jl is a package similar to HyperOpt.jl, but explicitly focusing on distributed (multi-processing) hyperparameter optimization by design.
Author ThummeTo
Github
Popularity
8 Stars
Updated Last
10 Months Ago
Started In
June 2023

What is DistributedHyperOpt.jl?

DistributedHyperOpt.jl is a package similar to HyperOpt.jl, but explicitly focusing on distributed (multi-processing) hyperparameter optimization by design.

Test (latest) Test (LTS)

How can I use DistributedHyperOpt.jl?

1. Open a Julia-REPL, switch to package mode using ], activate your preferred environment.

2. Install DistributedHyperOpt.jl:

(@v1.X) pkg> add "https://github.com/ThummeTo/DistributedHyperOpt.jl"

3. If you want to check that everything works correctly, you can run the tests bundled with DistributedHyperOpt.jl:

(@v1.X) pkg> test DistributedHyperOpt

4. See the folder examples (or testing scripts) for examples.

What is currently supported in DistributedHyperOpt.jl?

Max. processes
Random Sampler unlimited
Hyperband (using Random Sampler) num. brackets (s_max+1)