提交 88c192c8 编写于 作者: M Megvii Engine Team

feat(lite): add get_data_by_share in megenginelite python interface

GitOrigin-RevId: 0ddbb75e823106a61d5802d8e395db99a3e9f1d6
上级 8624ec22
......@@ -8,6 +8,7 @@
# "AS IS" BASIS, WITHOUT ARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
from .base import *
from .base import version as __version__
from .global_setting import *
from .network import *
from .struct import *
......
......@@ -69,7 +69,9 @@ class LiteOptions(Structure):
"const_shape": bool(self.const_shape),
"force_dynamic_alloc": bool(self.force_dynamic_alloc),
"force_output_dynamic_alloc": bool(self.force_output_dynamic_alloc),
"force_output_nocopy": bool(self.force_output_nocopy),
"force_output_use_user_specified_memory": bool(
self.force_output_use_user_specified_memory
),
"no_profiling_on_shape_change": bool(self.no_profiling_on_shape_change),
"jit_level": self.jit_level,
"comp_node_seq_record_level": self.comp_node_seq_record_level,
......@@ -99,7 +101,7 @@ class LiteConfig(Structure):
("device_id", c_int),
("device_type", c_int),
("backend", c_int),
("bare_model_cryption_name", c_char_p),
("_bare_model_cryption_name", c_char_p),
("options", LiteOptions),
]
......@@ -110,18 +112,30 @@ class LiteConfig(Structure):
else:
self.options = LiteOptions()
self.bare_model_cryption_name = c_char_p(b"")
self._bare_model_cryption_name = c_char_p(b"")
self.use_loader_dynamic_param = 0
self.has_compression = 0
self.backend = LiteBackend.LITE_DEFAULT
@property
def bare_model_cryption_name(self):
return self._bare_model_cryption_name.decode("utf-8")
@bare_model_cryption_name.setter
def bare_model_cryption_name(self, name):
if isinstance(name, str):
self._bare_model_cryption_name = name.encode("utf-8")
else:
assert isinstance(name, bytes), "name should be str or bytes type."
self._bare_model_cryption_name = name
def __repr__(self):
data = {
"has_compression": bool(self.has_compression),
"device_id": LiteDeviceType(self.device_id),
"device_type": LiteDeviceType(self.device_type),
"backend": LiteBackend(self.backend),
"bare_model_cryption_name": self.bare_model_cryption_name.decode("utf-8"),
"bare_model_cryption_name": self.bare_model_cryption_name,
"options": self.options,
}
return data.__repr__()
......@@ -149,7 +163,7 @@ class LiteIO(Structure):
"""
_fields_ = [
("name", c_char_p),
("_name", c_char_p),
("is_host", c_int),
("io_type", c_int),
("config_layout", LiteLayout),
......@@ -159,9 +173,9 @@ class LiteIO(Structure):
self, name, is_host=True, io_type=LiteIOType.LITE_IO_VALUE, layout=None
):
if type(name) == str:
self.name = c_char_p(name.encode("utf-8"))
self._name = c_char_p(name.encode("utf-8"))
else:
self.name = c_char_p(name)
self._name = c_char_p(name)
if layout:
self.config_layout = layout
......@@ -171,6 +185,18 @@ class LiteIO(Structure):
self.is_host = is_host
self.io_type = io_type
@property
def name(self):
return self._name.decode("utf-8")
@name.setter
def name(self, name):
if isinstance(name, str):
self._name = name.encode("utf-8")
else:
assert isinstance(name, bytes), "name should be str or bytes type."
self._name = name
def __repr__(self):
data = {
"name": self.name,
......@@ -208,17 +234,45 @@ class LiteNetworkIO(object):
the input and output information for user to construct _LiteNetWorkIO
"""
def __init__(self):
def __init__(self, inputs=None, outputs=None):
self.inputs = []
self.outputs = []
if inputs:
for i in inputs:
if isinstance(i, list):
self.inputs.append(LiteIO(*i))
else:
assert isinstance(
i, LiteIO
), "the param to construct LiteNetworkIO must be list of the LiteIO member or the LiteIO."
self.inputs.append(i)
if outputs:
for i in outputs:
if isinstance(i, list):
self.outputs.append(LiteIO(*i))
else:
assert isinstance(
i, LiteIO
), "the param to construct LiteNetworkIO must be list of the LiteIO member or the LiteIO."
self.outputs.append(i)
def add_input(self, input_io):
assert isinstance(input_io, LiteIO)
self.inputs.append(input_io)
def add_input(
self, obj, is_host=True, io_type=LiteIOType.LITE_IO_VALUE, layout=None
):
if isinstance(obj, LiteIO):
self.inputs.append(obj)
else:
name = obj
self.add_input(LiteIO(name, is_host, io_type, layout))
def add_output(self, output_io):
assert isinstance(output_io, LiteIO)
self.outputs.append(output_io)
def add_output(
self, obj, is_host=True, io_type=LiteIOType.LITE_IO_VALUE, layout=None
):
if isinstance(obj, LiteIO):
self.outputs.append(obj)
else:
name = obj
self.add_output(LiteIO(name, is_host, io_type, layout))
def _create_network_io(self):
network_io = _LiteNetworkIO()
......
......@@ -48,6 +48,15 @@ ctype_to_lite_dtypes = {
c_ushort: LiteDataType.LITE_UINT16,
}
_lite_dtypes_to_ctype = {
LiteDataType.LITE_INT: c_int,
LiteDataType.LITE_FLOAT: c_float,
LiteDataType.LITE_UINT8: c_ubyte,
LiteDataType.LITE_INT8: c_byte,
LiteDataType.LITE_INT16: c_short,
LiteDataType.LITE_UINT16: c_ushort,
}
class LiteLayout(Structure):
"""
......@@ -55,7 +64,7 @@ class LiteLayout(Structure):
"""
_fields_ = [
("shapes", c_size_t * MAX_DIM),
("_shapes", c_size_t * MAX_DIM),
("ndim", c_size_t),
("data_type", c_int),
]
......@@ -64,10 +73,10 @@ class LiteLayout(Structure):
if shape:
shape = list(shape)
assert len(shape) <= MAX_DIM, "Layout max dim is 7."
self.shapes = (c_size_t * MAX_DIM)(*shape)
self._shapes = (c_size_t * MAX_DIM)(*shape)
self.ndim = len(shape)
else:
self.shapes = (c_size_t * MAX_DIM)()
self._shapes = (c_size_t * MAX_DIM)()
self.ndim = 0
if not dtype:
self.data_type = LiteDataType.LITE_FLOAT
......@@ -83,9 +92,24 @@ class LiteLayout(Structure):
else:
raise RuntimeError("unkonw data type")
@property
def dtype(self):
return _lite_type_to_nptypes[LiteDataType(self.data_type)]
@property
def shapes(self):
return list(self._shapes)[0 : self.ndim]
@shapes.setter
def shapes(self, shape):
shape = list(shape)
assert len(shape) <= MAX_DIM, "Layout max dim is 7."
self._shapes = (c_size_t * MAX_DIM)(*shape)
self.ndim = len(shape)
def __repr__(self):
data = {
"shapes": list(self.shapes)[0 : self.ndim],
"shapes": self.shapes,
"ndim": self.ndim,
"data_type": _lite_type_to_nptypes[LiteDataType(self.data_type)],
}
......@@ -177,15 +201,20 @@ class LiteTensor(object):
device_type=LiteDeviceType.LITE_CPU,
device_id=0,
is_pinned_host=False,
shapes=None,
dtype=None,
):
"""
create a Tensor with layout, device, is_pinned_host param
create a Tensor with layout, device, is_pinned_host or shapes, dtype,
device_type, device_id, is_pinned_host param
"""
self._tensor = _Ctensor()
if layout:
self._layout = layout
else:
self._layout = LiteLayout()
if layout is not None:
self._layout = layout
elif shapes is not None:
shapes = list(shapes)
self._layout = LiteLayout(shapes, dtype)
self._device_type = device_type
self._device_id = device_id
self._is_pinned_host = is_pinned_host
......@@ -222,9 +251,12 @@ class LiteTensor(object):
@layout.setter
def layout(self, layout):
assert isinstance(layout, LiteLayout)
if isinstance(layout, LiteLayout):
self._layout = layout
self._api.LITE_set_tensor_layout(self._tensor, layout)
elif isinstance(layout, list):
self._layout.shapes = layout
self._api.LITE_set_tensor_layout(self._tensor, self._layout)
@property
def is_pinned_host(self):
......@@ -270,7 +302,6 @@ class LiteTensor(object):
"""
get the length of the meomry in byte
"""
self.update()
length = c_size_t()
self._api.LITE_get_tensor_total_size_in_byte(self._tensor, byref(length))
return length.value
......@@ -336,7 +367,6 @@ class LiteTensor(object):
"""
get the memory of the tensor, return c_void_p of the tensor memory
"""
self.update()
mem = c_void_p()
self._api.LITE_get_tensor_memory(self._tensor, byref(mem))
return mem
......@@ -347,7 +377,6 @@ class LiteTensor(object):
param data: the data will shared to the tensor, it should be a
numpy.ndarray or ctypes data
"""
self.update()
if isinstance(data, np.ndarray):
assert (
self.is_continue
......@@ -356,8 +385,7 @@ class LiteTensor(object):
self.is_pinned_host or self.device_type == LiteDeviceType.LITE_CPU
), "set_data_by_share can only apply in cpu tensor or pinned tensor."
np_type = _lite_type_to_nptypes[LiteDataType(self._layout.data_type)]
c_type = np.ctypeslib.as_ctypes_type(np_type)
c_type = _lite_dtypes_to_ctype[LiteDataType(self._layout.data_type)]
if self.nbytes != data.nbytes:
self.layout = LiteLayout(data.shape, ctype_to_lite_dtypes[c_type])
......@@ -377,7 +405,6 @@ class LiteTensor(object):
param data: the data to copy to tensor, it should be list,
numpy.ndarraya or ctypes with length
"""
self.update()
if layout is not None:
self.layout = layout
......@@ -386,8 +413,7 @@ class LiteTensor(object):
self.is_pinned_host or self.device_type == LiteDeviceType.LITE_CPU
), "set_data_by_copy can only apply in cpu tensor or pinned tensor."
np_type = _lite_type_to_nptypes[LiteDataType(self._layout.data_type)]
c_type = np.ctypeslib.as_ctypes_type(np_type)
c_type = _lite_dtypes_to_ctype[LiteDataType(self._layout.data_type)]
tensor_memory = c_void_p()
......@@ -415,6 +441,22 @@ class LiteTensor(object):
self._api.LITE_get_tensor_memory(self._tensor, byref(tensor_memory))
memmove(tensor_memory, data, data_length)
def get_data_by_share(self):
"""
get the data in the tensor, add share the data with a new numpy, and
return the numpy arrray, be careful, the data in numpy is valid before
the tensor memory is write again, such as LiteNetwok forward next time.
"""
assert self.is_continue, "get_data_by_share can only apply in continue tensor."
assert (
self.is_pinned_host or self.device_type == LiteDeviceType.LITE_CPU
), "get_data_by_share can only apply in CPU tensor or cpu pinned tensor."
memory = self.get_ctypes_memory()
c_type = _lite_dtypes_to_ctype[LiteDataType(self._layout.data_type)]
pnt = cast(memory, POINTER(c_type))
return np.ctypeslib.as_array(pnt, self._layout.shapes)
def to_numpy(self):
"""
get the buffer of the tensor
......@@ -475,3 +517,13 @@ def LiteTensorConcat(
)
result_tensor.update()
return result_tensor
def lite_dtype_2_numpy(dtype):
"""
convert lite dtype to corresponding numpy dtype
"""
assert isinstance(
dtype, LiteDataType
), "input must be LiteDataType when using lite_dtype_2_numpy."
return _lite_type_to_nptypes[dtype]
......@@ -21,6 +21,12 @@ def test_version():
print("Lite verson: {}".format(version))
def test_config():
config = LiteConfig()
config.bare_model_cryption_name = "nothing"
print(config)
def test_network_io():
input_io1 = LiteIO("data1", is_host=False, io_type=LiteIOType.LITE_IO_VALUE)
input_io2 = LiteIO(
......@@ -32,6 +38,7 @@ def test_network_io():
io = LiteNetworkIO()
io.add_input(input_io1)
io.add_input(input_io2)
io.add_input("data3", False)
output_io1 = LiteIO("out1", is_host=False)
output_io2 = LiteIO("out2", is_host=True, layout=LiteLayout([1, 1000]))
......@@ -39,7 +46,7 @@ def test_network_io():
io.add_output(output_io1)
io.add_output(output_io2)
assert len(io.inputs) == 2
assert len(io.inputs) == 3
assert len(io.outputs) == 2
assert io.inputs[0] == input_io1
......@@ -47,9 +54,25 @@ def test_network_io():
c_io = io._create_network_io()
assert c_io.input_size == 2
assert c_io.input_size == 3
assert c_io.output_size == 2
ins = [["data1", True], ["data2", False, LiteIOType.LITE_IO_SHAPE]]
outs = [["out1", True], ["out2", False, LiteIOType.LITE_IO_VALUE]]
io2 = LiteNetworkIO(ins, outs)
assert len(io2.inputs) == 2
assert len(io2.outputs) == 2
io3 = LiteNetworkIO([input_io1, input_io2], [output_io1, output_io2])
assert len(io3.inputs) == 2
assert len(io3.outputs) == 2
test_io = LiteIO("test")
assert test_io.name == "test"
test_io.name = "test2"
assert test_io.name == "test2"
class TestShuffleNet(unittest.TestCase):
source_dir = os.getenv("LITE_TEST_RESOURCE")
......@@ -319,9 +342,9 @@ class TestNetwork(TestShuffleNet):
data = ios[key].to_numpy().flatten()
input_data = self.input_data.flatten()
assert data.size == input_data.size
assert io.name.decode("utf-8") == "data"
assert io.name == "data"
for i in range(data.size):
assert data[i] == input_data[i]
assert abs(data[i] - input_data[i]) < 1e-5
return 0
network.set_start_callback(start_callback)
......@@ -343,7 +366,7 @@ class TestNetwork(TestShuffleNet):
output_data = self.correct_data.flatten()
assert data.size == output_data.size
for i in range(data.size):
assert data[i] == output_data[i]
assert abs(data[i] - output_data[i]) < 1e-5
return 0
network.set_finish_callback(finish_callback)
......@@ -404,3 +427,27 @@ class TestNetwork(TestShuffleNet):
binary_equal_between_batch=True,
)
self.do_forward(network)
def test_device_tensor_no_copy(self):
# construct LiteOption
net_config = LiteConfig()
net_config.options.force_output_use_user_specified_memory = True
network = LiteNetwork(config=net_config)
network.load(self.model_path)
input_tensor = network.get_io_tensor("data")
# fill input_data with device data
input_tensor.set_data_by_share(self.input_data)
output_tensor = network.get_io_tensor(network.get_output_name(0))
out_array = np.zeros(output_tensor.layout.shapes, output_tensor.layout.dtype)
output_tensor.set_data_by_share(out_array)
# inference
for i in range(2):
network.forward()
network.wait()
self.check_correct(out_array)
......@@ -54,6 +54,16 @@ def test_tensor_make():
tensor = LiteTensor(layout, device_id=1)
assert tensor.device_id == 1
tensor.layout = [8, 14]
assert tensor.layout.shapes[0] == 8
assert tensor.layout.shapes[1] == 14
assert tensor.layout.data_type == LiteDataType.LITE_FLOAT
tensor_new = LiteTensor(shapes=[1, 3, 224], dtype=np.int8)
assert tensor_new.layout.shapes[1] == 3
assert tensor_new.layout.shapes[2] == 224
assert tensor_new.layout.data_type == LiteDataType.LITE_INT8
def test_tensor_set_data():
layout = LiteLayout([2, 16], "int8")
......@@ -292,3 +302,24 @@ def test_tensor_concat():
for i in range(128):
index = j * 128 + i
assert real_data[index // 32][index % 32] == j
def test_tensor_get_memory_by_share():
layout = LiteLayout([4, 32], "int16")
tensor = LiteTensor(layout)
assert tensor.nbytes == 4 * 32 * 2
arr = np.ones([4, 32], "int16")
for i in range(128):
arr[i // 32][i % 32] = i
tensor.set_data_by_copy(arr)
test_data = tensor.get_data_by_share()
real_data = tensor.to_numpy()
for i in range(128):
assert real_data[i // 32][i % 32] == test_data[i // 32][i % 32]
arr[1][18] = 5
arr[3][7] = 345
tensor.set_data_by_copy(arr)
assert test_data[1][18] == 5
assert test_data[3][7] == 345
Markdown is supported
0% .
You are about to add 0 people to the discussion. Proceed with caution.
先完成此消息的编辑!
想要评论请 注册