This package creates a sparse matrix as Dictionary. You can convert the dictionary to a SparseCSC matrix or a dense matrix. It is useful when the nonzero elements randomly occur and you can not prepare the sparse storage before you see the actual elements.
The constructor of the matrix is SparseMatrixDict{Tv,Ti}(m,n) where Tv is the type of element (default:Float64), Ti is the type of index (default:Int), m is the number of rows, and n is the number of columns.
A pair of indices (row i and column j) will be treated as a tuple (i,j) and it is the key of dictionary; Dict{Tuple{Ti,Ti},Tv}.
using SparseArrays
using LinearAlgebra
using SparseMatrixDicts
n = 5
A = SparseMatrixDict(n,n) # default={Float64,Int}
# assignment; similar to regular matrices
A[1,1] = 2.0
A[2,5] = 1.0
# convert to dense matrix
dA = Matrix(A)
# convert to sparse matrix CSC
sA = sparse(A)
# make a symmetric sparse matrix for a square matrix
symA = Symmetric(sparse(A),:U) # :U for upper
You can use SparseMatrixCSC(A) instead of sparse(A).