InfiniteArrays.jl

A Julia package for representing infinite-dimensional arrays
Popularity
72 Stars
Updated Last
2 Months Ago
Started In
March 2018

InfiniteArrays.jl

Build Status codecov PkgEval

A Julia package for representing arrays with infinite dimension sizes, designed to work with other array types. Infinite arrays are by necessity lazy, and so this package is closely linked to LazyArrays.jl.

The package currently supports construction, but future versions will include support for linear algebra operations like matrix multiplication, LU decompositions, and QR decompositions.

Here are some examples of current functionality:

julia> using InfiniteArrays, LinearAlgebra, FillArrays

julia> x = Ones(∞) # infinite vector of all ones
Ones{Float64,1,Tuple{InfiniteArrays.OneToInf{Int64}}} with indices OneToInf():
 1.0
 1.0
 1.0
 1.0
 1.0
 1.0
 1.0
 1.0
 1.0
 1.0
 1.0
 1.0
 1.0
 

julia> cumsum(x) # infinite sum
1.0:1.0:+∞

julia> exp.(-(1:∞)) .+ 2 # broadcasting supported
BroadcastArray{Float64,1,Base.Broadcast.Broadcasted{LazyArrays.LazyArrayStyle{1},Tuple{InfiniteArrays.OneToInf{Int64}},typeof(+),Tuple{BroadcastArray{Float64,1,Base.Broadcast.Broadcasted{LazyArrays.LazyArrayStyle{1},Tuple{InfiniteArrays.OneToInf{Int64}},typeof(exp),Tuple{InfiniteArrays.InfStepRange{Int64,Int64}}}},Int64}}} with indices OneToInf():
 2.3678794411714423
 2.135335283236613
 2.049787068367864
 2.018315638888734
 2.0067379469990856
 2.0024787521766663
 2.0009118819655547
 2.0003354626279024
 2.0001234098040865
 2.0000453999297623
 2.0000167017007904
 2.000006144212353
 2.000002260329407
 2.000000831528719
 2.0000003059023204
 2.0000001125351745
 2.0000000413993773
 2.00000001522998
 2.0000000056027964
 2.0000000020611535
 2.000000000758256
 2.0000000002789466
 2.000000000102619
 2.000000000037751
 2.000000000013888
 2.0000000000051092
 

julia> Diagonal(1:∞) # combines well with Base arrays
Diagonal{Int64,InfiniteArrays.InfUnitRange{Int64}} with indices OneToInf()×OneToInf():
 1                                    
   2                                
     3                              
       4                            
         5                          
           6                          
             7                      
               8                    
                 9                  
                   10               
                      11              
                         12         
                            13      
                               14   
                                  15
                                     
                                   
                                   
                                   
                                   
                                     
                                   
                                   
                                   
                                   
                                     
                                                 

 julia> C = cache(Ones(∞)); C[1] = 3; C # Use cache to make mutable
LazyArrays.CachedArray{Float64,1,Array{Float64,1},Ones{Float64,1,Tuple{InfiniteArrays.InfiniteCardinal{0}}}} with indices OneToInf():
 3.0
 1.0
 1.0
 1.0
 1.0
 1.0
 1.0
 1.0
 1.0
 1.0
 1.0
 1.0
 1.0
 1.0
 1.0
 1.0
 1.0
 1.0
 1.0
 1.0
 1.0
 1.0
 1.0
 1.0
 1.0
 1.0