Skip to content

Open-source datasets for high-contrast imaging

License

Notifications You must be signed in to change notification settings

JuliaHCI/HCIDatasets.jl

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

39 Commits
 
 
 
 
 
 
 
 
 
 
 
 
 
 

Repository files navigation

HCIDatasets.jl

Build Status Coverage License

This repository provides a common location for high-contrast imaging datasets in an easy-to-consume interface.

Datasets

The following datasets are provided:

  • HR8799 - HR8799 taken with NIRC2/Vortex Coronagraph
  • BetaPictoris - Beta Pictoris taken with NaCo IRDIS
  • V471Tau - V 471 Tau IFS data cube taken with VLT/SPHERE

Usage

For the datasets provided, take a look at their docstrings to see a description and the available fields, for example

julia> ?HCIDatasets.HR8799

or directly query using keys

julia> keys(HCIDatasets.HR8799)
(:cube, :pa)

We can load the ADI cube and parallactic angles in this instance, just like a dictionary

julia> cube = HCIDatasets.HR8799[:cube]

julia> cube, angles = HCIDatasets.BetaPictoris[:cube, :pa]

and we can get the keys or iterate over the pairs

julia> using HCIDatasets: HR8799

julia> keys(HR8799)
(:cube, :pa)

julia> pairs(HR8799) |> collect
(:cube => Float32[ ... ],
 :pa => Float32[ ... ])

Data Dependencies

The datasets provided here are supplied via DataDeps.jl. The first time you try accessing a dataset you will prompted to download the necessary files. For more information about options for where this data is saved and how it is loaded, see the documentation.

Note: for package writers using this for testing with CI- DataDeps.jl can be set to automatically download via the environment variable DATADEPS_ALWAYS_ACCEPT=true, which will need to be set in your CI environment.

Adding a Dataset

If you would like to add a dataset, please upload it to zenodo and then file a pull request with the corresponding code additions. The code behind HCIDatasets.jl is very simple: browse the HR8799.jl file to get an idea of the process.

License

The code is under an MIT license (see LICENSE). Each dataset may have its own license or citations. Please provide attribution for the original data providers when making use of their data!