def test_subset_minmax(ctx_getter):
context = ctx_getter()
queue = cl.CommandQueue(context)
from pyopencl.clrandom import rand as clrand
l_a = 200000
gran = 5
l_m = l_a - l_a // gran + 1
if has_double_support():
dtypes = [numpy.float64, numpy.float32, numpy.int32]
else:
dtypes = [numpy.float32, numpy.int32]
for dtype in dtypes:
a_gpu = clrand(context, queue, (l_a,), dtype)
a = a_gpu.get()
meaningful_indices_gpu = cl_array.zeros(l_m, dtype=numpy.int32)
meaningful_indices = meaningful_indices_gpu.get()
j = 0
for i in range(len(meaningful_indices)):
meaningful_indices[i] = j
j = j + 1
if j % gran == 0:
j = j + 1
meaningful_indices_gpu = cl_array.to_device(meaningful_indices)
b = a[meaningful_indices]
min_a = numpy.min(b)
min_a_gpu = cl_array.subset_min(meaningful_indices_gpu, a_gpu).get()
assert min_a_gpu == min_a
def test_subset_minmax(ctx_factory):
from pytest import importorskip
importorskip("mako")
context = ctx_factory()
queue = cl.CommandQueue(context)
from pyopencl.clrandom import rand as clrand
l_a = 200000
gran = 5
l_m = l_a - l_a // gran + 1
if has_double_support(context.devices[0]):
dtypes = [np.float64, np.float32, np.int32]
else:
dtypes = [np.float32, np.int32]
for dtype in dtypes:
a_gpu = clrand(queue, (l_a,), dtype)
a = a_gpu.get()
meaningful_indices_gpu = cl_array.zeros(
queue, l_m, dtype=np.int32)
meaningful_indices = meaningful_indices_gpu.get()
j = 0
for i in range(len(meaningful_indices)):
meaningful_indices[i] = j
j = j + 1
if j % gran == 0:
j = j + 1
meaningful_indices_gpu = cl_array.to_device(
queue, meaningful_indices)
b = a[meaningful_indices]
min_a = np.min(b)
min_a_gpu = cl_array.subset_min(meaningful_indices_gpu, a_gpu).get()
assert min_a_gpu == min_a
def test_subset_minmax(ctx_factory):
from pytest import importorskip
importorskip("mako")
context = ctx_factory()
queue = cl.CommandQueue(context)
from pyopencl.clrandom import rand as clrand
l_a = 200000
gran = 5
l_m = l_a - l_a // gran + 1
if has_double_support(context.devices[0]):
dtypes = [np.float64, np.float32, np.int32]
else:
dtypes = [np.float32, np.int32]
for dtype in dtypes:
a_gpu = clrand(queue, (l_a,), dtype)
a = a_gpu.get()
meaningful_indices_gpu = cl_array.zeros(
queue, l_m, dtype=np.int32)
meaningful_indices = meaningful_indices_gpu.get()
j = 0
for i in range(len(meaningful_indices)):
meaningful_indices[i] = j
j = j + 1
if j % gran == 0:
j = j + 1
meaningful_indices_gpu = cl_array.to_device(
queue, meaningful_indices)
b = a[meaningful_indices]
min_a = np.min(b)
min_a_gpu = cl_array.subset_min(meaningful_indices_gpu, a_gpu).get()
assert min_a_gpu == min_a
def test_subset_minmax(ctx_getter):
context = ctx_getter()
queue = cl.CommandQueue(context)
from pyopencl.clrandom import rand as clrand
l_a = 200000
gran = 5
l_m = l_a - l_a // gran + 1
if has_double_support():
dtypes = [numpy.float64, numpy.float32, numpy.int32]
else:
dtypes = [numpy.float32, numpy.int32]
for dtype in dtypes:
a_gpu = clrand(context, queue, (l_a, ), dtype)
a = a_gpu.get()
meaningful_indices_gpu = cl_array.zeros(l_m, dtype=numpy.int32)
meaningful_indices = meaningful_indices_gpu.get()
j = 0
for i in range(len(meaningful_indices)):
meaningful_indices[i] = j
j = j + 1
if j % gran == 0:
j = j + 1
meaningful_indices_gpu = cl_array.to_device(meaningful_indices)
b = a[meaningful_indices]
min_a = numpy.min(b)
min_a_gpu = cl_array.subset_min(meaningful_indices_gpu,
a_gpu).get()
assert min_a_gpu == min_a
