def test_hostfromgpu_shape_i():
# Test that the shape is lifted over hostfromgpu
m = mode_with_gpu.including("local_dot_to_dot22",
"local_dot22_to_dot22scalar", "specialize")
a = tt.fmatrix("a")
ca = theano.gpuarray.type.GpuArrayType("float32", (False, False))()
av = np.asarray(np.random.rand(5, 4), dtype="float32")
cv = gpuarray.asarray(np.random.rand(5, 4),
dtype="float32",
context=get_context(test_ctx_name))
f = theano.function([a], GpuFromHost(test_ctx_name)(a), mode=m)
assert any(
isinstance(x.op, GpuFromHost) for x in f.maker.fgraph.toposort())
f = theano.function([a], GpuFromHost(test_ctx_name)(a).shape, mode=m)
topo = f.maker.fgraph.toposort()
assert isinstance(topo[0].op, tt.opt.Shape_i)
assert isinstance(topo[1].op, tt.opt.Shape_i)
assert isinstance(topo[2].op, tt.opt.MakeVector)
assert tuple(f(av)) == (5, 4)
f = theano.function([ca], host_from_gpu(ca), mode=m)
assert host_from_gpu in [x.op for x in f.maker.fgraph.toposort()]
f = theano.function([ca], host_from_gpu(ca).shape, mode=m)
topo = f.maker.fgraph.toposort()
assert isinstance(topo[0].op, theano.compile.Shape_i)
assert isinstance(topo[1].op, theano.compile.Shape_i)
assert isinstance(topo[2].op, tt.opt.MakeVector)
assert tuple(f(cv)) == (5, 4)
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 = theano.function([a], GpuFromHost(test_ctx_name)(a))
fv = f(av)
assert GpuArrayType.values_eq(fv, gv)
f = theano.function([g], host_from_gpu(g))
fv = f(gv)
assert np.all(fv == av)
def local_mypool_dnn_alternative(node):
if not dnn_available():
return
if isinstance(node.op, MyPool):
if not node.op.ignore_border:
return
img, = node.inputs
ds = node.op.ds
stride = node.op.st
pad = node.op.padding
mode = node.op.mode
if (img.owner and isinstance(img.owner.op, HostFromGpu)):
ret = dnn_pool(gpu_contiguous(img.owner.inputs[0]),
ds, stride=stride, pad=pad, mode=mode)
return [host_from_gpu(ret)]
def test_transfer_strided():
# This is just to ensure that it works in theano
# 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 = theano.function([a], GpuFromHost(test_ctx_name)(a))
fv = f(av)
assert GpuArrayType.values_eq(fv, gv)
f = theano.function([g], host_from_gpu(g))
fv = f(gv)
assert np.all(fv == av)
def local_mypool_dnn_grad_stride(node):
if not dnn_available():
return
if isinstance(node.op, MyMaxPoolGrad):
if not node.op.ignore_border:
return
inp, out, inp_grad = node.inputs
ds = node.op.ds
st = node.op.st
pad = node.op.padding
mode = node.op.mode
if ((inp.owner and isinstance(inp.owner.op, HostFromGpu)) or
(out.owner and isinstance(out.owner.op, HostFromGpu)) or
(inp_grad.owner and isinstance(inp_grad.owner.op,
HostFromGpu))):
ret = GpuDnnPoolGrad(mode=mode)(gpu_contiguous(inp),
gpu_contiguous(out),
gpu_contiguous(inp_grad),
ds, st, pad)
return [host_from_gpu(ret)]