Returns a sparse matrix of double-precision, floating-point values created from the individual coordinate format arrays, without any internal memory allocations.
- iOS 11+
- macOS 10.13+
- Mac Catalyst 13.0+
- tvOS 11+
- watchOS 4+
Number of rows in the structure.
Number of columns in the structure.
Number of blocks in the matrix.
Block size for data storage on both input and output.
Attributes of the matrix to create. The matrix will be forced to conform to the specified attributes by copying or dropping elements as needed.
Row indices of the matrix structure.
Column indices of the matrix structure.
Contents of the structurally nonzero (block) matrix elements.
A block of memory to store the returned matrix.
Workspace of size
Count * sizeof(int)
Out-of-range entries are dropped and duplicate entries are summed.
For symmetric matrices, entries are accepted in either triangle. If they are in the "wrong" triangle, as specified by
triangle, they are transposed. If an entry is already present, they are treated as duplicates and summed.
For triangular matrices, entries in the "wrong" triangle as specified by
triangle are treated as out-of-range and are dropped.