提交 ec9a55ae 编写于 作者: Y Yiqun Liu 提交者: GitHub

Merge pull request #3927 from Xreki/core_add_fc_op

Port fully connected operator
/* Copyright (c) 2016 PaddlePaddle Authors. All Rights Reserve.
Licensed under the Apache License, Version 2.0 (the "License");
you may not use this file except in compliance with the License.
You may obtain a copy of the License at
http://www.apache.org/licenses/LICENSE-2.0
Unless required by applicable law or agreed to in writing, software
distributed under the License is distributed on an "AS IS" BASIS,
WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
See the License for the specific language governing permissions and
limitations under the License. */
#include "paddle/framework/op_registry.h"
#include "paddle/operators/net_op.h"
namespace paddle {
namespace operators {
class FCOp : public NetOp {
public:
FCOp(const std::string &type, const framework::VariableNameMap &inputs,
const framework::VariableNameMap &outputs,
const framework::AttributeMap &attrs)
: NetOp(type, inputs, outputs, attrs) {
PADDLE_ENFORCE(!Inputs("X").empty(),
"Inputs(X) of FCOp should not be null.");
PADDLE_ENFORCE(!Inputs("W").empty(),
"Inputs(W) of FCOp should not be null.");
PADDLE_ENFORCE(!Outputs("MulOut").empty(),
"Outputs(MulOut) of FCOp should not be null.");
PADDLE_ENFORCE_NE(Output("Out"), framework::kEmptyVarName,
"Output(Out) of FCOp should not be null.");
auto x = Inputs("X");
auto w = Inputs("W");
auto mul_out = Outputs("MulOut");
PADDLE_ENFORCE_EQ(
x.size(), w.size(),
"The size of inputs X(%d) should be the same as that of weights W(%d).",
x.size(), w.size());
PADDLE_ENFORCE_EQ(mul_out.size(), x.size(),
"The size of intermediate mul_out(%d) should be the same "
"as that of inputs X(%d).",
mul_out.size(), x.size());
size_t n = x.size();
PADDLE_ENFORCE_GE(n, static_cast<size_t>(1),
"The size of inputs X(%d) should be no less than 1.", n);
auto x_num_col_dims = Attr<std::vector<int>>("xNumColDims");
// Set all values or set no values (use the default value)
if (!x_num_col_dims.empty()) {
PADDLE_ENFORCE_EQ(x_num_col_dims.size(), n,
"The size of attribute xNumColDims(%d) should be the "
"same as that of inputs X(%d).",
x_num_col_dims.size(), n);
} else {
x_num_col_dims.resize(n);
for (size_t i = 0; i < n; i++) {
x_num_col_dims[i] = 1;
}
}
// mul_out[i] = X[i] * W[i]
for (size_t i = 0; i < n; i++) {
framework::AttributeMap mul_attr;
mul_attr["x_num_col_dims"] = static_cast<int>(x_num_col_dims[i]);
mul_attr["y_num_col_dims"] = static_cast<int>(1);
AppendOp(
framework::OpRegistry::CreateOp("mul", {{"X", {x[i]}}, {"Y", {w[i]}}},
{{"Out", {mul_out[i]}}}, mul_attr));
}
// sum_out = X[0] * W[0] + ... + X[n-1] * W[n-1]
auto sum_out = mul_out[0];
if (n > 1) {
PADDLE_ENFORCE_NE(Output("SumOut"), framework::kEmptyVarName,
"Output(SumOut) of FCOp should not be null when the "
"size of Inputs(X) > 1.");
sum_out = Output("SumOut");
AppendOp(framework::OpRegistry::CreateOp("sum", {{"X", {mul_out}}},
{{"Out", {sum_out}}}, {}));
} else {
if (Output("SumOut") != framework::kEmptyVarName) {
this->Rename(Output("SumOut"), framework::kEmptyVarName);
}
}
// add_out = sum_out + b
auto b = Input("B");
auto add_out = sum_out;
if (b != framework::kEmptyVarName) {
PADDLE_ENFORCE_NE(
Output("AddOut"), framework::kEmptyVarName,
"Output(AddOut) of FCOp should not be null when Input(B) is set.");
add_out = Output("AddOut");
AppendOp(framework::OpRegistry::CreateOp(
"rowwise_add", {{"X", {sum_out}}, {"b", {Input("B")}}},
{{"Out", {add_out}}}, {}));
} else {
if (Output("AddOut") != framework::kEmptyVarName) {
this->Rename(Output("AddOut"), framework::kEmptyVarName);
}
}
auto activation = Attr<std::string>("activation");
AppendOp(framework::OpRegistry::CreateOp(activation, {{"X", {add_out}}},
{{"Y", {Output("Out")}}}, {}));
CompleteAddOp(false);
}
};
class FCOpMaker : public framework::OpProtoAndCheckerMaker {
public:
FCOpMaker(framework::OpProto *proto, framework::OpAttrChecker *op_checker)
: OpProtoAndCheckerMaker(proto, op_checker) {
AddInput("X",
"(A vector of Tensors) each input Tensor can be of arbitrary "
"dimension, and will be reshaped to a 2-D matrix of size "
"(minibatch, number_of_input_features) according to attribute "
"xNumColDims.")
.AsDuplicable();
AddInput("W",
"(A vector of Tensors) the weights of FC operator, a "
"vector of 2-D matrix of size "
"(number_of_input_features, number_of_neurons).")
.AsDuplicable();
AddInput("B",
"(Tensor) the bias of FC operator, a 1-D vector of size "
"number_of_neurons.");
AddOutput("Out",
"(Tensor) the activated output matrix of FC operator, a 2-D "
"matrix of size (minibatch, number_of_neurons).");
AddOutput("MulOut",
"(A vector of Tensors) the intermediate outputs of FC operator, "
"each Tensor saving the product of X_i * W_i.")
.AsIntermediate()
.AsDuplicable();
AddOutput(
"SumOut",
"(Tensor) the intermediate output of FC operator, "
"saving the sum of the products of X and W, that is sum{X_i * W_i}.")
.AsIntermediate();
AddOutput("AddOut",
"(Tensor) the non-actived output of FC operator, "
"saving sum{X_i * W_i} + B.")
.AsIntermediate();
AddAttr<std::string>(
"activation",
"(string, default identity) the activation type of FC operator.")
.SetDefault("identity")
.InEnum({"identity", "sigmoid", "softmax"});
AddAttr<std::vector<int>>(
"xNumColDims",
"(std::vector<int>) The inputs Tensors of FC operator can be of "
"more than 2 dimensions. In that case, each input Tensor `X_i` will be "
"reshaped to a 2-D matrix. The matrix's first dimension "
"(the length of column) will be the product of `X_i`'s last "
"`xNumColDims_i` dimensions, that is "
"`X_i.dims[0] x ... x X_i.dims[xNumColDims_i - 1]`. "
"The matrix's second dimension (the length of row) will be the product "
"of `X_i`'s first `rank - xNumColDims_i` dimensions, that is "
"`X_i.dims[xNumColDims_i] x ... x X_i.dims[rank - 1]`)")
.SetDefault(std::vector<int>{});
AddComment(R"DOC(
Fully Connected Operator, known as Fully Connected Layer or Inner Product Layer
in Convolutional Neural Networks. Neurons in a fully connected layer have
full connections to all activations in the previous layer.
It computes an inner product of a set of
learned weights with a matrix multiplication followed by a bias offset
(optionally).
Equation:
Out = Act(sum_n{X_i * W_i} + B)
where X_i is Tensor that will be reshaped to a 2-D matrix of size (M x K),
usually M is the minibatch size and K is the number of input features.
W_i is a 2-D matrix of size (K x N), where N means the number of neurons
in the fully connected layer. B is a 1-D vector of size N.
Thus, the output Out is a 2-D matrix of size (M x N).
Activation type can be set to `identity` (default), `sigmoid` or `softmax`.
)DOC");
}
};
} // namespace operators
} // namespace paddle
namespace ops = paddle::operators;
REGISTER_OP_WITHOUT_GRADIENT(fc, ops::FCOp, ops::FCOpMaker);
...@@ -27,7 +27,7 @@ class IdentityOpMaker : public framework::OpProtoAndCheckerMaker { ...@@ -27,7 +27,7 @@ class IdentityOpMaker : public framework::OpProtoAndCheckerMaker {
framework::OpAttrChecker *op_checker) framework::OpAttrChecker *op_checker)
: OpProtoAndCheckerMaker(proto, op_checker) { : OpProtoAndCheckerMaker(proto, op_checker) {
AddInput("X", "The input tensor of identity operator."); AddInput("X", "The input tensor of identity operator.");
AddOutput("Out", "The output tensor of identity operator."); AddOutput("Y", "The output tensor of identity operator.");
AddComment(R"DOC( AddComment(R"DOC(
The identity operator is an alias of the scale operator The identity operator is an alias of the scale operator
with the attribute scale fixed to 1.0. with the attribute scale fixed to 1.0.
...@@ -44,12 +44,13 @@ class IdentityOp : public NetOp { ...@@ -44,12 +44,13 @@ class IdentityOp : public NetOp {
: NetOp(type, inputs, outputs, attrs) { : NetOp(type, inputs, outputs, attrs) {
PADDLE_ENFORCE_NE(Input("X"), framework::kEmptyVarName, PADDLE_ENFORCE_NE(Input("X"), framework::kEmptyVarName,
"Input(X) of IdentityOp should not be null."); "Input(X) of IdentityOp should not be null.");
PADDLE_ENFORCE_NE(Output("Out"), framework::kEmptyVarName, PADDLE_ENFORCE_NE(Output("Y"), framework::kEmptyVarName,
"Output(Out) of IdentityOp should not be null."); "Output(Y) of IdentityOp should not be null.");
AppendOp(framework::OpRegistry::CreateOp( AppendOp(framework::OpRegistry::CreateOp(
"scale", {{"X", {Input("X")}}}, {{"Out", {Output("Out")}}}, "scale", {{"X", {Input("X")}}}, {{"Out", {Output("Y")}}},
{{"scale", static_cast<AttrType>(1)}})); {{"scale", static_cast<AttrType>(1)}}));
CompleteAddOp(false);
} }
}; };
......
...@@ -71,7 +71,7 @@ class MinusGradOp : public NetOp { ...@@ -71,7 +71,7 @@ class MinusGradOp : public NetOp {
// x_grad = out_grad // x_grad = out_grad
AppendOp(framework::OpRegistry::CreateOp("identity", {{"X", {out_grad}}}, AppendOp(framework::OpRegistry::CreateOp("identity", {{"X", {out_grad}}},
{{"Out", {x_grad}}}, {})); {{"Y", {x_grad}}}, {}));
framework::AttributeMap scale_attr; framework::AttributeMap scale_attr;
scale_attr["scale"] = static_cast<AttrType>(-1); scale_attr["scale"] = static_cast<AttrType>(-1);
......
if(WITH_PYTHON) if(WITH_PYTHON)
cc_library(paddle_pybind SHARED cc_library(paddle_pybind SHARED
SRCS pybind.cc SRCS pybind.cc
DEPS pybind python backward DEPS pybind python backward
${GLOB_OP_LIB}) ${GLOB_OP_LIB})
......
...@@ -28,10 +28,10 @@ def create_op(scope, op_type, inputs, outputs, attrs): ...@@ -28,10 +28,10 @@ def create_op(scope, op_type, inputs, outputs, attrs):
if out_name in outputs: if out_name in outputs:
kwargs[out_name] = [] kwargs[out_name] = []
if out_dup: if out_dup:
sub_in = outputs[out_name] sub_out = outputs[out_name]
for sub_in_name, _ in sub_in: for sub_out_name, _ in sub_out:
var = scope.new_var(sub_in_name) var = scope.new_var(sub_out_name)
kwargs[out_name].append(sub_in_name) kwargs[out_name].append(sub_out_name)
else: else:
var = scope.new_var(out_name) var = scope.new_var(out_name)
kwargs[out_name].append(out_name) kwargs[out_name].append(out_name)
...@@ -39,6 +39,7 @@ def create_op(scope, op_type, inputs, outputs, attrs): ...@@ -39,6 +39,7 @@ def create_op(scope, op_type, inputs, outputs, attrs):
for attr_name in Operator.get_op_attr_names(op_type): for attr_name in Operator.get_op_attr_names(op_type):
if attr_name in attrs: if attr_name in attrs:
kwargs[attr_name] = attrs[attr_name] kwargs[attr_name] = attrs[attr_name]
return Operator(op_type, **kwargs) return Operator(op_type, **kwargs)
...@@ -179,8 +180,9 @@ class OpTest(unittest.TestCase): ...@@ -179,8 +180,9 @@ class OpTest(unittest.TestCase):
def check_output_with_place(self, place): def check_output_with_place(self, place):
self.scope = core.Scope() self.scope = core.Scope()
op_inputs = self.inputs if hasattr(self, "inputs") else dict() op_inputs = self.inputs if hasattr(self, "inputs") else dict()
op_outputs = self.outputs if hasattr(self, "outputs") else dict()
op_attrs = self.attrs if hasattr(self, "attrs") else dict() op_attrs = self.attrs if hasattr(self, "attrs") else dict()
self.op = create_op(self.scope, self.op_type, op_inputs, self.outputs, self.op = create_op(self.scope, self.op_type, op_inputs, op_outputs,
op_attrs) op_attrs)
if isinstance(place, core.GPUPlace) and not self.op.support_gpu(): if isinstance(place, core.GPUPlace) and not self.op.support_gpu():
return return
...@@ -192,21 +194,23 @@ class OpTest(unittest.TestCase): ...@@ -192,21 +194,23 @@ class OpTest(unittest.TestCase):
for out_name, out_dup in Operator.get_op_outputs(self.op.type()): for out_name, out_dup in Operator.get_op_outputs(self.op.type()):
if out_dup: if out_dup:
sub_out = self.outputs[out_name] sub_out = self.outputs[out_name]
for sub_out_name in sub_out: for sub_out_name, sub_out_array in sub_out:
actual = np.array( actual = np.array(
self.scope.find_var(sub_out_name).get_tensor()) self.scope.find_var(sub_out_name).get_tensor())
expect = sub_out[sub_out_name] expect = sub_out_array
self.assertTrue( self.assertTrue(
np.allclose( np.allclose(
actual, expect, atol=1e-05), actual, expect, atol=1e-05),
"output name: " + out_name + "has diff") "output name: " + out_name + " has diff")
else: else:
actual = np.array(self.scope.find_var(out_name).get_tensor()) var = self.scope.find_var(out_name)
expect = self.outputs[out_name] if var is not None:
self.assertTrue( actual = np.array(var.get_tensor())
np.allclose( expect = self.outputs[out_name]
actual, expect, atol=1e-05), self.assertTrue(
"output name: " + out_name + "has diff") np.allclose(
actual, expect, atol=1e-05),
"output name: " + out_name + " has diff")
def check_output(self): def check_output(self):
places = [core.CPUPlace()] places = [core.CPUPlace()]
...@@ -241,8 +245,9 @@ class OpTest(unittest.TestCase): ...@@ -241,8 +245,9 @@ class OpTest(unittest.TestCase):
max_relative_error=0.005): max_relative_error=0.005):
self.scope = core.Scope() self.scope = core.Scope()
op_inputs = self.inputs if hasattr(self, "inputs") else dict() op_inputs = self.inputs if hasattr(self, "inputs") else dict()
op_outputs = self.outputs if hasattr(self, "outputs") else dict()
op_attrs = self.attrs if hasattr(self, "attrs") else dict() op_attrs = self.attrs if hasattr(self, "attrs") else dict()
self.op = create_op(self.scope, self.op_type, op_inputs, self.outputs, self.op = create_op(self.scope, self.op_type, op_inputs, op_outputs,
op_attrs) op_attrs)
if no_grad_set is None: if no_grad_set is None:
no_grad_set = set() no_grad_set = set()
......
import unittest
import numpy as np
from op_test import OpTest
class TestFCOp1(OpTest):
def setUp(self):
x0 = np.random.random((16, 32)).astype("float32")
w0 = np.random.random((32, 10)).astype("float32")
mul_out0 = np.dot(x0, w0)
identity_out = mul_out0
self.op_type = "fc"
self.inputs = {"X": [("X0", x0)], "W": [("W0", w0)]}
self.outputs = {"MulOut": [("MulOut0", mul_out0)], "Out": identity_out}
def test_check_output(self):
self.check_output()
def test_check_grad(self):
self.check_grad(["X0", "W0"], "Out", max_relative_error=0.01)
class TestFCOp2(OpTest):
def setUp(self):
x0 = np.random.random((16, 4, 8)).astype("float32")
x1 = np.random.random((4, 4, 32)).astype("float32")
w0 = np.random.random((32, 10)).astype("float32")
w1 = np.random.random((32, 10)).astype("float32")
b = np.random.random(10).astype("float32")
mul_out0 = np.dot(x0.reshape(16, 4 * 8), w0)
mul_out1 = np.dot(x1.reshape(4 * 4, 32), w1)
sum_out = mul_out0 + mul_out1
add_out = np.add(sum_out, b)
sigmoid_out = 1 / (1 + np.exp(-add_out))
self.op_type = "fc"
self.inputs = {
"X": [("X0", x0), ("X1", x1)],
"W": [("W0", w0), ("W1", w1)],
"B": b
}
self.attrs = {"xNumColDims": [1, 2], "activation": "sigmoid"}
self.outputs = {
"MulOut": [("MulOut0", mul_out0), ("MulOut1", mul_out1)],
"SumOut": sum_out,
"AddOut": add_out,
"Out": sigmoid_out
}
def test_check_output(self):
self.check_output()
def test_check_grad(self):
self.check_grad(
["X0", "X1", "W0", "W1", "B"], "Out", max_relative_error=0.01)
if __name__ == '__main__':
unittest.main()
...@@ -7,13 +7,13 @@ class TestIdentityOp(OpTest): ...@@ -7,13 +7,13 @@ class TestIdentityOp(OpTest):
def setUp(self): def setUp(self):
self.op_type = "identity" self.op_type = "identity"
self.inputs = {'X': np.random.random((10, 10)).astype("float32")} self.inputs = {'X': np.random.random((10, 10)).astype("float32")}
self.outputs = {'Out': self.inputs['X']} self.outputs = {'Y': self.inputs['X']}
def test_check_output(self): def test_check_output(self):
self.check_output() self.check_output()
def test_check_grad(self): def test_check_grad(self):
self.check_grad(['X'], 'Out') self.check_grad(['X'], 'Y')
if __name__ == "__main__": if __name__ == "__main__":
......
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