8000 ENH: handle empty matrices in qr decomposition by convexset · Pull Request #11593 · numpy/numpy · GitHub
[go: up one dir, main page]
Skip to content

ENH: handle empty matrices in qr decomposition #11593

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Merged
merged 12 commits into from
Jul 31, 2018
Merged

ENH: handle empty matrices in qr decomposition #11593

Changes from 1 commit
Commits
Show all changes
12 commits
Select commit Hold shift + click to select a range
36fcb0c
handle empty matrices in qr decomposition
convexset Jul 20, 2018
3fa693a
fixed offending quotes
convexset Jul 20, 2018
6dabc39
release notes updated
convexset Jul 20, 2018
448b409
revised to use existing array sizes; skipping calls to lapack in the …
convexset Jul 29, 2018
1624a80
setting things back where they were and calling LAPACK functions better
convexset Jul 29, 2018
e6fa7df
ensure lwork respects requirements of lapack methods (zgeqrf, dgeqrf,…
convexset Jul 29, 2018
7f22736
updated tests
convexset Jul 29, 2018
68d608e
simplified tests
convexset Jul 29, 2018
5809e9a
removed redundant test; expressed sub-test with pytest.mark.parametrize
convexset Jul 31, 2018
16d79e9
fixed dumb copy-and-paste error
convexset Jul 31, 2018
c4c6426
clarified terminology
convexset Jul 31, 2018
ff9063c
trimmed excess on line
convexset Jul 31, 2018
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
Prev Previous commit
Next Next commit
updated tests
  • Loading branch information
@convexset
convexset committed Jul 29, 2018
commit 7f2273674ed8b0082c6da12d15ce23151a1056f2
44 changes: 19 additions & 25 deletions numpy/linalg/tests/test_linalg.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -1583,45 +1583,39 @@ def check_qr(self, a):
assert_almost_equal(r2, r1)

def test_qr_empty(self):
for m, n in [(3, 0), (0, 3)]:
for m, n in [(0, 0), (3, 0), (0, 3)]:
k = min(m, n)
a = np.empty((m, n))
a_type = type(a)
a_dtype = a.dtype
#

q, r = np.linalg.qr(a, mode='reduced')
assert_(q.dtype == a_dtype)
assert_(r.dtype == a_dtype)
assert_equal(q.dtype, a_dtype)
assert_equal(r.dtype, a_dtype)
assert_(isinstance(q, a_type))
assert_(isinstance(r, a_type))
assert_(q.shape == (m, k))
assert_(r.shape == (k, n))
assert_almost_equal(np.dot(q, r), a)
assert_almost_equal(np.dot(q.T.conj(), q), np.eye(k))
assert_almost_equal(np.triu(r), r)
#
assert_equal(q.shape, (m, k))
assert_equal(r.shape, (k, n))

r = np.linalg.qr(a, mode='r')
Copy link
Member

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Isn't this already covered by check_qr?

assert_(r.dtype == a_dtype)
assert_equal(r.dtype, a_dtype)
assert_(isinstance(r, a_type))
assert_(r.shape == (k, n))
assert_almost_equal(np.triu(r), r)
#
assert_equal(r.shape, (k, n))
Copy link
Member

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

This shape test should probably be in check_qr anyway

Copy link
Contributor Author
@convexset convexset Jul 31, 2018 •
edited
Loading

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

It's there implicitly.

        # mode == 'reduced'
        q1, r1 = linalg.qr(a, mode='reduced')
        # ...
        assert_(r1.shape == (k, n))
        # ...

        # mode == 'r'
        r2 = linalg.qr(a, mode='r')
        # ...
        assert_almost_equal(r2, r1)

So I took that bit out.

Copy link
Member

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Fair enough. I think the explicit shape check you had here was a little clearer, but it's not super important.


q, r = np.linalg.qr(a, mode='complete')
assert_(q.dtype == a_dtype)
assert_(r.dtype == a_dtype)
assert_equal(q.dtype, a_dtype)
assert_equal(r.dtype, a_dtype)
assert_(isinstance(q, a_type))
assert_(isinstance(r, a_type))
assert_(q.shape == (m, m))
assert_(r.shape == (m, n))
assert_equal(q.shape, (m, m))
assert_equal(r.shape, (m, n))
assert_almost_equal(q, np.eye(m))
assert_almost_equal(np.dot(q, r), a)
assert_almost_equal(np.triu(r), r)
#

h, tau = np.linalg.qr(a, mode='raw')
Copy link
Member

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

This test can stay - for whatever reason, it seems check_qr decided not to test it

assert_(h.dtype == np.double)
assert_(tau.dtype == np.double)
assert_(h.shape == (n, m))
assert_(tau.shape == (k,))
assert_equal(h.dtype, np.double)
assert_equal(tau.dtype, np.double)
assert_equal(h.shape, (n, m))
assert_equal(tau.shape, (k,))

def test_mode_raw(self):
# The factorization is not unique and varies between libraries,
Expand Down
0