Fixed norm test for every mode and fixed test for old numpy version.

theano/sandbox/linalg/ops.py

@@ -905,9 +905,7 @@ eig = Eig()
 
 class SVD(Op):
     """
-    Singular Value Decomposition.
-    Factors the matrix a as u * np.diag(s) * v, where u and v are unitary
-        and s is a 1-d array of a's singular values.
+    See doc in the docstring of the function just after this class.
     """
     _numop = staticmethod(numpy.linalg.svd)
 
@@ -1097,7 +1095,6 @@ def qr(a, mode="full"):
     x = [[2, 1], [3, 4]]
     if isinstance(numpy.linalg.qr(x,mode), tuple):
         return QRFull(mode)(a)
-
     else:
         return QRIncomplete(mode)(a)
 

theano/sandbox/linalg/tests/test_linalg.py

@@ -176,60 +176,32 @@ def test_matrix_dot():
     assert _allclose(numpy_sol, theano_sol)
 
 
-def test_qr_default():
-    rng = numpy.random.RandomState(utt.fetch_seed())
-    A = tensor.matrix("A", dtype=theano.config.floatX)
-    Q = qr(A)
-    fn = function([A], Q)
-    a = rng.rand(4, 4).astype(theano.config.floatX)
-    n_q, n_r = numpy.linalg.qr(a)
-    t_q, t_r = fn(a)
-
-    assert _allclose(n_q, t_q)
-    assert _allclose(n_r, t_r)
-
-
 def test_qr_modes():
     rng = numpy.random.RandomState(utt.fetch_seed())
-    A = tensor.matrix("A", dtype=theano.config.floatX)
-    Q = qr(A, mode="reduced")
-    R = qr(A, mode="complete")
-    S = qr(A, mode="r")
-    T = qr(A, mode="raw")
-    U = qr(A, mode="full")
-    V = qr(A, mode="economic")
-    fq = function([A], Q)
-    fr = function([A], R)
-    fs = function([A], S)
-    ft = function([A], T)
-    fu = function([A], U)
-    fv = function([A], V)
 
+    A = tensor.matrix("A", dtype=theano.config.floatX)
     a = rng.rand(4, 4).astype(theano.config.floatX)
 
-    n_q = numpy.linalg.qr(a, mode="reduced")
-    t_q = fq(a)
-
-    n_r = numpy.linalg.qr(a, mode="complete")
-    t_r = fr(a)
-
-    n_s = numpy.linalg.qr(a, mode="r")
-    t_s = fs(a)
-
-    n_t = numpy.linalg.qr(a, mode="raw")
-    t_t = ft(a)
-
-    n_u = numpy.linalg.qr(a, mode="full")
-    t_u = fu(a)
-
-    n_v = numpy.linalg.qr(a, mode="economic")
-    t_v = fv(a)
-
-    assert _allclose(n_q, t_q)
-    assert _allclose(n_r, t_r)
-    assert _allclose(n_s, t_s)
-    assert _allclose(n_u, t_u)
-    assert _allclose(n_v, t_v)
+    try:
+        numpy.linalg.qr(a, "complete")
+    except TypeError, e:
+        assert "name 'complete' is not defined" in str(e)
+        raise SkipTest
+
+    f = function([A], qr(A))
+    t_qr = f(a)
+    n_qr = numpy.linalg.qr(a)
+    assert _allclose(n_qr, t_qr)
+
+    for mode in ["reduced", "complete", "r", "raw", "full", "economic"]:
+        f = function([A], qr(A, mode))
+        t_qr = f(a)
+        n_qr = numpy.linalg.qr(a, mode)
+        if isinstance(n_qr, (list, tuple)):
+            assert _allclose(n_qr[0], t_qr[0])
+            assert _allclose(n_qr[1], t_qr[1])
+        else:
+            assert _allclose(n_qr, t_qr)
 
 
 def test_svd():