def test_misc(init_pyDive):
sizes = ((10,20,30), (30,20,10), (16,16,16), (16,32,48), (13,29,37))
def do_funny_stuff(a, b):
a[1:,1:,1:] = b[:-1,:-1,:-1]
b[1:,1:,1:] = a[1:,:-1,:-1]
a[1:,1:,1:] = b[1:,1:,:-1]
b[1:,1:,1:] = a[1:,:-1,1:]
a[:-1,:-3,:-4] = b[1:,3:,4:]
b[0,:,0] += a[-2,:,-2]
a[4:-3,2:-1,5:-2] = b[4:-3,2:-1,5:-2]
b[1:3,1:4,1:5] *= a[-3:-1,-4:-1,-5:-1]
a[1,2,3] -= b[3,2,1]
b[:,:,0] = a[:,:,1]
for size in sizes:
print "size", size
np_a = (np.random.rand(*size) * 100.0).astype(np.int)
np_b = (np.random.rand(*size) * 100.0).astype(np.int)
pd_a = pyDive.array(np_a, distaxes=(0,1,2))
pd_b = pyDive.array(np_b, distaxes=(0,1,2))
do_funny_stuff(np_a, np_b)
do_funny_stuff(pd_a, pd_b)
assert np.array_equal(pd_a, np_a)
assert np.array_equal(pd_b, np_b)
def test_interengine(init_pyDive):
for size in sizes:
for dtype in dtypes:
ref = (np.random.rand(*size) * 100.0).astype(dtype)
test_array_cpu = pyDive.array(ref)
test_array = pyDive.gpu.empty(size, dtype, distaxes=range(len(size)))
test_array[:] = test_array_cpu
for i in range(len(size)):
if size[i] < 5: continue
slicesA = [slice(None)] * len(size)
slicesB = list(slicesA)
slicesA[i] = slice(0, 5)
slicesB[i] = slice(-5, None)
test_array[slicesA] = test_array[slicesB]
ref[slicesA] = ref[slicesB]
assert np.array_equal(test_array.to_cpu().gather(), ref)
slicesA = [s/2 for s in size]
slicesB = list(slicesA)
slicesA[i] = slice(0, 5)
slicesB[i] = slice(-5, None)
test_array[slicesA] = test_array[slicesB]
ref[slicesA] = ref[slicesB]
assert np.array_equal(test_array.to_cpu().gather(), ref)
def test_basics(init_pyDive):
for size in sizes:
for dtype in dtypes:
ref = (np.random.rand(*size) * 100.0).astype(dtype)
test_array_cpu = pyDive.array(ref)
test_array = pyDive.gpu.empty(size, dtype, distaxes=range(len(size)))
test_array[:] = test_array_cpu
if all(s > 3 for s in size):
slices = [slice(1, -2, None) for i in range(len(size))]
a = test_array[slices]
a = a + 1
a = a**2
a = a + a
test_array[slices] = a
b = ref[slices]
b = b + 1
b = b**2
b = b + b
ref[slices] = b
np.testing.assert_array_almost_equal(test_array.to_cpu().gather(), ref)
def test_interengine(init_pyDive):
for size in sizes:
for dtype in dtypes:
ref = (np.random.rand(*size) * 100.0).astype(dtype)
test_array_cpu = pyDive.array(ref)
test_array = pyDive.gpu.empty(size,
dtype,
distaxes=range(len(size)))
test_array[:] = test_array_cpu
for i in range(len(size)):
if size[i] < 5: continue
slicesA = [slice(None)] * len(size)
slicesB = list(slicesA)
slicesA[i] = slice(0, 5)
slicesB[i] = slice(-5, None)
test_array[slicesA] = test_array[slicesB]
ref[slicesA] = ref[slicesB]
assert np.array_equal(test_array.to_cpu().gather(), ref)
slicesA = [s / 2 for s in size]
slicesB = list(slicesA)
slicesA[i] = slice(0, 5)
slicesB[i] = slice(-5, None)
test_array[slicesA] = test_array[slicesB]
ref[slicesA] = ref[slicesB]
assert np.array_equal(test_array.to_cpu().gather(), ref)
def test_basics(init_pyDive):
for size in sizes:
for dtype in dtypes:
ref = (np.random.rand(*size) * 100.0).astype(dtype)
test_array_cpu = pyDive.array(ref)
test_array = pyDive.gpu.empty(size,
dtype,
distaxes=range(len(size)))
test_array[:] = test_array_cpu
if all(s > 3 for s in size):
slices = [slice(1, -2, None) for i in range(len(size))]
a = test_array[slices]
a = a + 1
a = a**2
a = a + a
test_array[slices] = a
b = ref[slices]
b = b + 1
b = b**2
b = b + b
ref[slices] = b
np.testing.assert_array_almost_equal(
test_array.to_cpu().gather(), ref)
def test_slicing(init_pyDive):
for size in sizes:
for dtype in dtypes:
ref = (np.random.rand(*size) * 100.0).astype(dtype)
for distaxis in range(len(size)):
test_array = pyDive.empty(size, dtype, distaxis)
test_array[:] = ref
slices = []
for i in range(len(size)):
start = size[i] / 3
stop = size[i] - size[i] / 5
step = 2
slices.append(slice(start, stop, step))
assert np.array_equal(ref[slices], test_array[slices].gather())
slices = []
for i in range(len(size)):
slices.append(slice(-5, None, None))
assert np.array_equal(ref[slices], test_array[slices].gather())
slices = []
for i in range(len(size)):
slices.append(slice(0, 5, None))
assert np.array_equal(ref[slices], test_array[slices].gather())
# bitmask indexing
bitmask = pyDive.array(np.random.rand(*size) > 0.5, distaxes=test_array.distaxes)
assert ref[bitmask.gather()].shape == test_array[bitmask].shape
# ordering can be distinct, thus merely check if sets are equal
assert set(ref[bitmask.gather()]) == set(test_array[bitmask].gather())
ref2 = ref.copy()
test_array2 = test_array.copy()
ref2[bitmask.gather()] = 1
test_array2[bitmask] = 1
assert np.array_equal(ref2, test_array2.gather())
