Skip to content

yngtdd/hyperspace

Repository files navigation


DOI

Welcome to HyperSpace!

If you have a complicated model with many hyperparameters, there is a lot to explore here. A combinatorial explosion, in fact. Details can be found in the documentation.

Here is a Hitchhiker's Guide:

Preparations

You can't just waltz out into space without the proper preparations!

HyperSpace makes use of MPI through mpi4py. Make sure to have either MPICH or Open MPI.

git clone https://github.com/yngtodd/hyperspace.git
cd hyperspace

# Get you gear!
pip install .

Modules

Here is a Hubble height view of the library:

Space

"Space," it says, "is big. Really big. You just won't believe how vastly, hugely, mindbogglingly big it is. I mean, you may think it's a long way down the road to the chemist's, but that's just peanuts to space." - The Hitchhiker's Guide to the Galaxy

In space you will find the various classes that define hyperparameter search spaces.

Mapping Space

In mapping_space we have functions that define hyperspaces, the many subregions of our hyperparameter search space to be distributed across cluster resources.

Hyperdrive

In hyperdrive we have various methods for distributing our optimization procedure.

@misc{hyperspace,
  author = {M.Todd Young},
  title = {HyperSpace},
  year = {2017},
  publisher = {GitHub},
  journal = {GitHub repository},
  howpublished = {\url{https://github.com/yngtodd/hyperspace}},
}