def test_rebroadcast():
for dtype in ["float16", "float32"]:
a = rand_gpuarray(1, dtype=dtype)
g = GpuArrayType(dtype=dtype, broadcastable=(False,))("g")
f = aesara.function([g], Rebroadcast((0, True))(g))
assert isinstance(f.maker.fgraph.toposort()[0].op, Rebroadcast)
res = f(a)
assert GpuArrayType.values_eq(res, a)
def test_deep_copy():
for dtype in ["float16", "float32"]:
a = rand_gpuarray(20, dtype=dtype)
g = GpuArrayType(dtype=dtype, broadcastable=(False,))("g")
f = aesara.function([g], g)
assert isinstance(f.maker.fgraph.toposort()[0].op, DeepCopyOp)
res = f(a)
assert GpuArrayType.values_eq(res, a)
def test_view():
for dtype in ["float16", "float32"]:
a = rand_gpuarray(20, dtype=dtype)
g = GpuArrayType(dtype=dtype, broadcastable=(False,))("g")
m = aesara.compile.get_default_mode().excluding("local_view_op")
f = aesara.function([g], ViewOp()(g), mode=m)
assert isinstance(f.maker.fgraph.toposort()[0].op, ViewOp)
res = f(a)
assert GpuArrayType.values_eq(res, a)
def test_transfer_cpu_gpu():
a = tt.fmatrix("a")
g = GpuArrayType(dtype="float32", broadcastable=(False, False))("g")
av = np.asarray(rng.rand(5, 4), dtype="float32")
gv = gpuarray.array(av, context=get_context(test_ctx_name))
f = aesara.function([a], GpuFromHost(test_ctx_name)(a))
fv = f(av)
assert GpuArrayType.values_eq(fv, gv)
f = aesara.function([g], host_from_gpu(g))
fv = f(gv)
assert np.all(fv == av)
def test_transfer_gpu_gpu():
g = GpuArrayType(dtype="float32",
broadcastable=(False, False),
context_name=test_ctx_name)()
av = np.asarray(rng.rand(5, 4), dtype="float32")
gv = gpuarray.array(av, context=get_context(test_ctx_name))
mode = mode_with_gpu.excluding("cut_gpua_host_transfers",
"local_cut_gpua_host_gpua")
f = aesara.function([g], GpuToGpu(test_ctx_name)(g), mode=mode)
topo = f.maker.fgraph.toposort()
assert len(topo) == 1
assert isinstance(topo[0].op, GpuToGpu)
fv = f(gv)
assert GpuArrayType.values_eq(fv, gv)
def test_transfer_strided():
# This is just to ensure that it works in aesara
# libgpuarray has a much more comprehensive suit of tests to
# ensure correctness
a = tt.fmatrix("a")
g = GpuArrayType(dtype="float32", broadcastable=(False, False))("g")
av = np.asarray(rng.rand(5, 8), dtype="float32")
gv = gpuarray.array(av, context=get_context(test_ctx_name))
av = av[:, ::2]
gv = gv[:, ::2]
f = aesara.function([a], GpuFromHost(test_ctx_name)(a))
fv = f(av)
assert GpuArrayType.values_eq(fv, gv)
f = aesara.function([g], host_from_gpu(g))
fv = f(gv)
assert np.all(fv == av)
