AxisIndices.jl

Here are some reasons you should try AxisIndices

• Flexible design for customizing multidimensional indexing behavior
• It's fast. StaticRanges are used to speed up indexing ranges. If something is slow, please create a detailed issue.
• Works with Julia's standard library (in progress). The end goal of AxisIndices is to fully integrate with the standard library wherever possible. If you can find a relevant method that isn't supported in Baseor Statistics then it's likely an oversight, so make an issue. LinearAlgebra, MappedArrays, OffsetArrays, and NamedDims also have some form of support.

The linked documentation provides a very brief "Quick Start" section along with detailed documentation of internal methods and types.

Construction By Axes

Axes can indicate what kind of array you want (static/fixed size) and/or can map key values to indices.

julia> using AxisIndices

julia> x = AxisArray{Int}(undef, OneToSRange(2), OneToSRange(2));

julia> x[1:4] .= 1;

julia> parent(x)
2×2 StaticArrays.MArray{Tuple{2,2},Int64,2,4} with indices SOneTo(2)×SOneTo(2):
 1  1
 1  1

julia> A = AxisArray(reshape(1:4, 2, 2), [:a, :b], ["one", "two"])
2×2 AxisArray{Int64,2}
 • dim_1 - [:a, :b]
 • dim_2 - ["one", "two"]
      one   two
  a     1     3
  b     2     4

julia> A[:a, "one"]
1

Indexing Cheat Sheet

The following can be replicated by using Unitful seconds (i.e., using Unitful: s)