COLAMDStats¶

class sksparse.colamd.COLAMDStats(N_rows_ignored: int, N_cols_ignored: int, Ncmpa: int, status: int, info1: int, info2: int, info3: int)[source]¶

Information statistics returned by the COLAMD algorithm.

This class wraps the contents of the stats array returned by C colamd() into a Python dataclass.

N_rows_ignored[source]¶

The number of dense or empty rows ignored in the ordering.

Type:: int

N_cols_ignored[source]¶

The number of dense or empty columns ignored in the ordering.

Type:: int

Ncmpa[source]¶

The number of garbage collections performed.

Type:: int

status[source]¶

Status code indicating the result of the COLAMD operation. If non-zero, colamd will throw an appropriate exception that interprets this status code.

Type:: int

The following fields take on different meanings depending on the value of

``status``

info1[source]¶

Value of status:

0: the highest numbered column that is unsorted or has duplicate entries.
-3: the value of n_row.
-4: the value of n_col.
-5: the value of nnz == p[n_col].
-6: the value of p[0].
-7: the required Alen value.
-8: the column with negative entries.
-9: the column with a row index out of bounds.

Type:: int

info2[source]¶

Value of status:

0: the last seen duplicate or unsorted row index.
-7: the actual Alen value.
-9: the bad row index.

Type:: int

info3[source]¶

Value of status:

0: the number of duplicates or unsorted row indices.
-9: n_row.

Type:: int

Notes

Field descriptions are adapted from SuiteSparse colamd.c [1].

Added in version 0.5.0.

References

Methods

`__init__`
`from_array`	Create a COLAMDStats instance from an array.