Revision history [back]

If principal submatrix of order k means intersection of the first k rows and first k columns, here is an example.

sage: m = matrix(ZZ, 4, range(4^2))
sage: m
[ 0  1  2  3]
[ 4  5  6  7]
[ 8  9 10 11]
[12 13 14 15]
sage: p = [m[0:k, 0:k] for k in range(m.ncols() + 1)]
sage: for a in p:
....:     print a
....:     print ''
....: 
[]

[0]

[0 1]
[4 5]

[ 0  1  2]
[ 4  5  6]
[ 8  9 10]

[ 0  1  2  3]
[ 4  5  6  7]
[ 8  9 10 11]
[12 13 14 15]

If principal submatrix of order k means intersection of the first k rows and first k columns, here is an example.

sage: m = matrix(ZZ, 4, range(4^2))
sage: m
[ 0  1  2  3]
[ 4  5  6  7]
[ 8  9 10 11]
[12 13 14 15]
sage: p = [m[0:k, 0:k] for k in range(m.ncols() + 1)]
sage: for a in p:
....:     print a
....:     print ''
....: 
[]

[0]

[0 1]
[4 5]

[ 0  1  2]
[ 4  5  6]
[ 8  9 10]

[ 0  1  2  3]
[ 4  5  6  7]
[ 8  9 10 11]
[12 13 14 15]

sage: def principal_submatrix(m, k):
....:     """
....:     Return the principal submatrix of m of order k
....:     """
....:     return m[0:k, 0:k]

sage: m = matrix(ZZ, 4, range(4^2))
sage: m
[ 0  1  2  3]
[ 4  5  6  7]
[ 8  9 10 11]
[12 13 14 15]
sage: principal_submatrix(m, 3)
[ 0  1  2]
[ 4  5  6]
[ 8  9 10]
sage: principal_submatrix(m, 2)
[0 1]
[4 5]

If The submatrix section of Wikipedia's matrix page mentions several possible definitions of principal submatrix of order k means intersection of the first k.

Given your question, you likely adopt the definition by which a principal submatrix of order k is a k by k submatrix in which the set of row indices that remain is the same as the set of column indices that remain.

So let us define principal_submatrix as a function that takes a matrix and a set of indices to keep.

def principal_submatrix(m, s, sort=False):
    """
    Return the submatrix of m corresponding to indices in s

    INPUT:

    - ``m`` -- a matrix

    - ``s`` -- an iterable for the indices of rows and columns to keep

    OUTPUT:

    The submatrix of ``m`` using the indices in ``s``.

    An optional argument specifies whether to sort the indices.
    Note that duplicate indices will be kept.
    """
    if sort:
        s = sorted(s)
    return m[s, s]

From there define principal_submatrices as a function that takes a matrix and the number of indices to keep.

def principal_submatrices(m, k):
    """
    Return the list of principal submatrices of m of order k

    These are the submatrices of m (assumed to be a square matrix)
    obtained by keeping k rows and first k columns,
here is an example.columns, with the same indices.
    """
    S = Subsets(range(m.ncols()), k)
    return [principal_submatrix(m, s, sort=True) for s in S]

To illustrate these functions, let us use the following 4 by 4 matrix.

sage: m = matrix(ZZ, 4, range(4^2))
sage: m
[ 0  1  2  3]
[ 4  5  6  7]
[ 8  9 10 11]
[12 13 14 15]
sage: p = [m[0:k, 0:k] for k in range(m.ncols() + 1)]
sage: for a in p:
....:     print a
....:     print ''
....: 
[]

[0]

Extract a specified principal submatrix:

sage: principal_submatrix(m, [0, 1, 3])
[ 0  1  3]
[ 4  5  7]
[12 13 15]

List all principal submatrices of each order:

sage: principal_submatrices(m, 0)
[[]]
sage: principal_submatrices(m, 1)
[[0], [5], [10], [15]]
sage: principal_submatrices(m, 2)
[
[0 1]
1]  [ 0  2]  [ 0  3]  [ 5  6]  [ 5  7]  [10 11]
[4 5]

5], [ 8 10], [12 15], [ 9 10], [13 15], [14 15]
]
sage: principal_submatrices(m, 3)
[
[ 0  1  2]
2]  [ 0  1  3]  [ 0  2  3]  [ 5  6  7]
[ 4  5  6]
6]  [ 4  5  7]  [ 8  10 11]  [ 9 10]

10 11]
[ 8  9 10], [12 13 15], [12 14 15], [13 14 15]
]
sage: principal_submatrices(m, 4)
[
[ 0  1  2  3]
[ 4  5  6  7]
[ 8  9 10 11]
[12 13 14 15]
]

sage: def principal_submatrix(m, k):
....:     """
....:     Return the principal submatrix of m of order k
....:     """
....:     return m[0:k, 0:k]

sage: m = matrix(ZZ, 4, range(4^2))
sage: m
[ 0  1  2  3]
[ 4  5  6  7]
[ 8  9 10 11]
[12 13 14 15]
sage: principal_submatrix(m, 3)
[ 0  1  2]
[ 4  5  6]
[ 8  9 10]
sage: principal_submatrix(m, 2)
[0 1]
[4 5]