def test_eye(self):
def get_offset(tensor_array):
x, y = tensor_array.shape
for k in range(x):
for l in range(y):
if tensor_array[k][l] == 1:
return k, l
return x, y
shape = 5
eye = ht.eye(shape, dtype=ht.uint8, split=1, device=ht_device)
self.assertIsInstance(eye, ht.DNDarray)
self.assertEqual(eye.dtype, ht.uint8)
self.assertEqual(eye.shape, (shape, shape))
self.assertEqual(eye.split, 1)
offset_x, offset_y = get_offset(eye._DNDarray__array)
self.assertGreaterEqual(offset_x, 0)
self.assertGreaterEqual(offset_y, 0)
x, y = eye._DNDarray__array.shape
for i in range(x):
for j in range(y):
expected = 1 if i - offset_x is j - offset_y else 0
self.assertEqual(eye._DNDarray__array[i][j], expected)
shape = (10, 20)
eye = ht.eye(shape, dtype=ht.float32, device=ht_device)
self.assertIsInstance(eye, ht.DNDarray)
self.assertEqual(eye.dtype, ht.float32)
self.assertEqual(eye.shape, shape)
self.assertEqual(eye.split, None)
offset_x, offset_y = get_offset(eye._DNDarray__array)
self.assertGreaterEqual(offset_x, 0)
self.assertGreaterEqual(offset_y, 0)
x, y = eye._DNDarray__array.shape
for i in range(x):
for j in range(y):
expected = 1.0 if i - offset_x is j - offset_y else 0.0
self.assertEqual(eye._DNDarray__array[i][j], expected)
shape = (10,)
eye = ht.eye(shape, dtype=ht.int32, split=0, device=ht_device)
self.assertIsInstance(eye, ht.DNDarray)
self.assertEqual(eye.dtype, ht.int32)
self.assertEqual(eye.shape, shape * 2)
self.assertEqual(eye.split, 0)
offset_x, offset_y = get_offset(eye._DNDarray__array)
self.assertGreaterEqual(offset_x, 0)
self.assertGreaterEqual(offset_y, 0)
x, y = eye._DNDarray__array.shape
for i in range(x):
for j in range(y):
expected = 1 if i - offset_x is j - offset_y else 0
self.assertEqual(eye._DNDarray__array[i][j], expected)
def test_qr_sp1_ext(self):
st_whole = torch.randn(70, 70, device=self.device.torch_device)
sp = 1
for m in range(50, st_whole.shape[0] + 1, 1):
for n in range(50, st_whole.shape[1] + 1, 1):
for t in range(1, 3):
st = st_whole[:m, :n].clone()
a_comp = ht.array(st, split=0)
a = ht.array(st, split=sp)
qr = a.qr(tiles_per_proc=t)
self.assertTrue(ht.allclose(a_comp, qr.Q @ qr.R, rtol=1e-5, atol=1e-5))
self.assertTrue(ht.allclose(qr.Q.T @ qr.Q, ht.eye(m), rtol=1e-5, atol=1e-5))
self.assertTrue(ht.allclose(ht.eye(m), qr.Q @ qr.Q.T, rtol=1e-5, atol=1e-5))
def test_fill_diagonal(self):
ref = ht.zeros((ht.MPI_WORLD.size * 2, ht.MPI_WORLD.size * 2),
dtype=ht.float32,
split=0)
a = ht.eye(ht.MPI_WORLD.size * 2, dtype=ht.float32, split=0)
a.fill_diagonal(0)
self.assertTrue(ht.equal(a, ref))
ref = ht.zeros((ht.MPI_WORLD.size * 2, ht.MPI_WORLD.size * 2),
dtype=ht.int32,
split=0)
a = ht.eye(ht.MPI_WORLD.size * 2, dtype=ht.int32, split=0)
a.fill_diagonal(0)
self.assertTrue(ht.equal(a, ref))
ref = ht.zeros((ht.MPI_WORLD.size * 2, ht.MPI_WORLD.size * 2),
dtype=ht.float32,
split=1)
a = ht.eye(ht.MPI_WORLD.size * 2, dtype=ht.float32, split=1)
a.fill_diagonal(0)
self.assertTrue(ht.equal(a, ref))
ref = ht.zeros((ht.MPI_WORLD.size * 2, ht.MPI_WORLD.size * 3),
dtype=ht.float32,
split=0)
a = ht.eye((ht.MPI_WORLD.size * 2, ht.MPI_WORLD.size * 3),
dtype=ht.float32,
split=0)
a.fill_diagonal(0)
self.assertTrue(ht.equal(a, ref))
# ToDo: uneven tensor dimensions x and y when bug in factories.eye is fixed
ref = ht.zeros((ht.MPI_WORLD.size * 3, ht.MPI_WORLD.size * 3),
dtype=ht.float32,
split=1)
a = ht.eye((ht.MPI_WORLD.size * 3, ht.MPI_WORLD.size * 3),
dtype=ht.float32,
split=1)
a.fill_diagonal(0)
self.assertTrue(ht.equal(a, ref))
# ToDo: uneven tensor dimensions x and y when bug in factories.eye is fixed
ref = ht.zeros((ht.MPI_WORLD.size * 4, ht.MPI_WORLD.size * 4),
dtype=ht.float32,
split=0)
a = ht.eye((ht.MPI_WORLD.size * 4, ht.MPI_WORLD.size * 4),
dtype=ht.float32,
split=0)
a.fill_diagonal(0)
self.assertTrue(ht.equal(a, ref))
a = ht.ones((ht.MPI_WORLD.size * 2, ), dtype=ht.float32, split=0)
with self.assertRaises(ValueError):
a.fill_diagonal(0)
def test_qr(self):
m, n = 20, 40
st = torch.randn(m,
n,
device=self.device.torch_device,
dtype=torch.float)
a_comp = ht.array(st, split=0)
for t in range(1, 3):
for sp in range(2):
a = ht.array(st, split=sp, dtype=torch.float)
qr = a.qr(tiles_per_proc=t)
self.assertTrue(
ht.allclose((a_comp - (qr.Q @ qr.R)),
0,
rtol=1e-5,
atol=1e-5))
self.assertTrue(
ht.allclose(qr.Q.T @ qr.Q, ht.eye(m), rtol=1e-5,
atol=1e-5))
self.assertTrue(
ht.allclose(ht.eye(m), qr.Q @ qr.Q.T, rtol=1e-5,
atol=1e-5))
m, n = 40, 40
st1 = torch.randn(m, n, device=self.device.torch_device)
a_comp1 = ht.array(st1, split=0)
for t in range(1, 3):
for sp in range(2):
a1 = ht.array(st1, split=sp)
qr1 = a1.qr(tiles_per_proc=t)
self.assertTrue(
ht.allclose((a_comp1 - (qr1.Q @ qr1.R)),
0,
rtol=1e-5,
atol=1e-5))
self.assertTrue(
ht.allclose(qr1.Q.T @ qr1.Q,
ht.eye(m),
rtol=1e-5,
atol=1e-5))
self.assertTrue(
ht.allclose(ht.eye(m),
qr1.Q @ qr1.Q.T,
rtol=1e-5,
atol=1e-5))
m, n = 40, 20
st2 = torch.randn(m,
n,
dtype=torch.double,
device=self.device.torch_device)
a_comp2 = ht.array(st2, split=0, dtype=ht.double)
for t in range(1, 3):
for sp in range(2):
a2 = ht.array(st2, split=sp)
qr2 = a2.qr(tiles_per_proc=t)
self.assertTrue(
ht.allclose(a_comp2, qr2.Q @ qr2.R, rtol=1e-5, atol=1e-5))
self.assertTrue(
ht.allclose(qr2.Q.T @ qr2.Q,
ht.eye(m, dtype=ht.double),
rtol=1e-5,
atol=1e-5))
self.assertTrue(
ht.allclose(ht.eye(m, dtype=ht.double),
qr2.Q @ qr2.Q.T,
rtol=1e-5,
atol=1e-5))
# test if calc R alone works
qr = ht.qr(a2, calc_q=False, overwrite_a=True)
self.assertTrue(qr.Q is None)
m, n = 40, 20
st = torch.randn(m, n, device=self.device.torch_device)
a_comp = ht.array(st, split=None)
a = ht.array(st, split=None)
qr = a.qr()
self.assertTrue(ht.allclose(a_comp, qr.Q @ qr.R, rtol=1e-5, atol=1e-5))
self.assertTrue(
ht.allclose(qr.Q.T @ qr.Q, ht.eye(m), rtol=1e-5, atol=1e-5))
self.assertTrue(
ht.allclose(ht.eye(m), qr.Q @ qr.Q.T, rtol=1e-5, atol=1e-5))
# raises
with self.assertRaises(TypeError):
ht.qr(np.zeros((10, 10)))
with self.assertRaises(TypeError):
ht.qr(a_comp, tiles_per_proc="ls")
with self.assertRaises(TypeError):
ht.qr(a_comp, tiles_per_proc=1, calc_q=30)
with self.assertRaises(TypeError):
ht.qr(a_comp, tiles_per_proc=1, overwrite_a=30)
with self.assertRaises(ValueError):
ht.qr(a_comp, tiles_per_proc=torch.tensor([1, 2, 3]))
with self.assertRaises(ValueError):
ht.qr(ht.zeros((3, 4, 5)))
