Skip to content

P
PaddleDetection

PaddlePaddle / PaddleDetection
大约 1 年 前同步成功

通知 694
Star 11112
Fork 2696
P
PaddleDetection

体验新版 GitCode，发现更多精彩内容 >>

提交 c7c25067 编写于 作者: T tensor-tang
浏览文件
操作

add forward implementation

上级 954b0e11
隐藏空白更改
内联 并排
Showing with 83 addition and 235 deletion
paddle/fluid/operators/fusion_seq_concat_fc_op.cc
浏览文件 @ c7c25067
......@@ -25,30 +25,15 @@ namespace operators {
void FusionSeqConcatFCOp::InferShape(framework::InferShapeContext* ctx) const {
PADDLE_ENFORCE(ctx->HasInput("X"),
"Input(X) of FusionSeqConcatFC should not be null.");
PADDLE_ENFORCE(ctx->HasInput("C0"),
"Input(C0) of FusionSeqConcatFC should not be null.");
PADDLE_ENFORCE(ctx->HasInput("LSTMWeight"),
"Input(LSTMWeight) of FusionSeqConcatFC should not be null.");
PADDLE_ENFORCE(ctx->HasInput("LSTMBias"),
"Input(LSTMBias) of FusionSeqConcatFC should not be null.");
PADDLE_ENFORCE(
ctx->HasInput("AttentionWeight"),
"Input(AttentionWeight) of FusionSeqConcatFC should not be null.");
PADDLE_ENFORCE(ctx->HasOutput("Hidden"),
"Output(Hidden) of FusionSeqConcatFC should not be null.");
PADDLE_ENFORCE(ctx->HasOutput("Cell"),
"Output(Cell) of FusionSeqConcatFC should not be null.");
PADDLE_ENFORCE(
ctx->HasOutput("AttentionedX"),
"Output(AttentionedX) of FusionSeqConcatFC should not be null.");
PADDLE_ENFORCE(
ctx->HasOutput("AttentionFCOut"),
"Output(AttentionFCOut) of FusionSeqConcatFC should not be null.");
PADDLE_ENFORCE(ctx->HasOutput("LSTMX"),
"Output(LSTMX) of FusionSeqConcatFC should not be null.");
PADDLE_ENFORCE(ctx->HasOutput("LSTMOUT"),
"Output(LSTMOUT) of FusionSeqConcatFC should not be null.");
PADDLE_ENFORCE(ctx->HasInput("FCWeight"),
"Input(FCWeight) of FusionSeqConcatFC should not be null.");
PADDLE_ENFORCE(ctx->HasOutput("Out"),
"Output(Out) of FusionSeqConcatFC should not be null.");
PADDLE_ENFORCE(ctx->HasOutput("FCOut"),
"Output(FCOut) of FusionSeqConcatFC should not be null.");
// need check fc height = all inputs width sum
auto x_dims = ctx->GetInputDim("X");
const int M = x_dims[1];
......@@ -120,6 +105,9 @@ void FusionSeqConcatFCOp::InferShape(framework::InferShapeContext* ctx) const {
// AttentionFCOut should be reshape as (maxseqlen,1) in runtime
ctx->ShareLoD("X", "Hidden");
ctx->ShareLoD("X", "Cell");
ctx->SetOutputDim("Out", out_dims);
ctx->ShareLoD("X", /*->*/ "Out");
}
framework::OpKernelType FusionSeqConcatFCOp::GetExpectedKernelType(
......@@ -131,95 +119,37 @@ framework::OpKernelType FusionSeqConcatFCOp::GetExpectedKernelType(
void FusionSeqConcatFCOpMaker::Make() {
AddInput("X",
"(LoDTensor) the input is a LodTensor, which support "
"variable-time length input sequence. The underlying tensor in "
"this LoDTensor is a matrix with shape (T X M), where T is the "
"total time steps in this mini-batch, M is the dim size of x.");
AddInput("C0",
"(Tensor) LSTM C0"
"This is a tensor with shape (N x D), where N is the batch size, D "
"is the gate size."
"C0 is necessary because of attention.");
AddInput("H0",
"(Tensor, optional) LSTM H0"
"This is a tensor with shape (N x D), where N is the "
"batch size and D is the gate size.")
.AsDispensable();
AddInput("AttentionWeight",
"(Tensor) the weights of attention fc. Always relu the fc result."
"The shape is ((M+D) x 1), where M is the dim size of x, D is the "
"gate size of LSTM.");
AddInput("AttentionBias",
"(Tensor, optional) the bias of attention fc."
"The shape is (1 x 1)")
.AsDispensable();
AddInput("AttentionScalar",
"(Tensor, optional) the scalar on the result of attentioned fc. "
"Always relu the Scalar."
"The shape is (1 x 1)")
.AsDispensable();
AddInput("AttentionScalarBias",
"(Tensor, optional) the scalar bias of attention fc."
"The shape is (1 x 1)")
.AsDispensable();
AddInput("LSTMWeight",
"(Tensor) the combined weight of LSTM"
" - The shape is ((D+M) x 4D), where D is the hidden gate size, M "
"is the dim size of x"
" - Weight = {W_forget, W_input, W_output, W_cell}");
AddInput("LSTMBias",
"(Tensor) the combined bias of LSTM, shape (1x4D)."
"Note: we should add the bias of hidden and context accorindg to "
"the same gate: "
"{B_forget, B_input, B_output, B_cell}");
AddOutput("Hidden",
"(LoDTensor) (same as LSTMOp) the hidden state of LSTM operator. "
"The shape is (T x D), and lod is the same with the `Input`.");
AddOutput("Cell",
"(LoDTensor) (same as LSTMOp) the cell state of LSTM operator. "
"The shape is (T x D), and lod is the same with the `Input`.");
AddOutput("AttentionedX",
"(Tensor) shape is (T x 1), the result after X * AttentionWeight,"
" where T is the total time steps in this mini-batch,"
" D is the hidden size.")
.AsIntermediate();
AddOutput("AttentionFCOut",
"(Tensor) (max_seq_len, 1), compute at each step.")
.AsIntermediate();
AddOutput("LSTMX",
"(Tensor) the input X of LSTM for each step."
"Shape is (1 x M), where M is the x frame size")
.AsIntermediate();
"(LoDTensor) input LodDTensors, the first one must be have ref lod "
"for sequence expand, and the rest input should have same lod.")
.AsDuplicable();
AddInput("FCWeight", "(Tensor) the weights of fc.");
AddInput("FCBias", "(Tensor, optional) the bias of fc.").AsDispensable();
AddOutput("Out", "(LoDTensor) Output LodTensor.");
AddOutput(
"LSTMOUT",
"(Tensor) the output of LSTM X(1*(D+M))* weight((D+M)*4D) for each step."
"Shape is (1 x 4D), where M is the x frame size")
"FCOut",
"(Tensor) the intermediate tensor to keep the result of fc."
"Shape is (N x D), where N is the batch size, D is the output dim of fc")
.AsIntermediate();
AddAttr<std::string>("gate_activation",
"(string, default: sigmoid)"
"The activation for input gate, forget gate and output "
"gate, `sigmoid` by default.")
.SetDefault("sigmoid")
.InEnum({"sigmoid", "tanh", "relu", "identity"});
AddAttr<std::string>("cell_activation",
"(string, default: tanh)"
"The activation for cell output, `tanh` by defalut.")
.SetDefault("tanh")
.InEnum({"sigmoid", "tanh", "relu", "identity"});
AddAttr<std::string>("candidate_activation",
"(string, default: tanh)"
"The activation for candidate hidden state, "
"`tanh` by default.")
.SetDefault("tanh")
AddAttr<std::string>("fc_activation",
"(string, default: identity)"
"The activation for the result of fc."
"`identity` by default.")
.SetDefault("identity")
.InEnum({"sigmoid", "tanh", "relu", "identity"});
AddComment(R"DOC(
Fusion Sequence expand + concat + fc Operator.
Only support seq_expand ref_level=0,
All below conditions should be meet:
and the ref lod of seq_expand level is the first input of concat,
The ref_level of seq_expand should be 0.
and the other inputs should have same lod and same batch size of ref lod.
The ref lod of seq_expand level is the first input of concat.
The other inputs should have same lod and same batch size of ref lod.
The seq len of other inputs should be 1.
The concat axis should be 1.
)DOC");
}
......@@ -257,150 +187,68 @@ class FusionSeqConcatFCKernel : public framework::OpKernel<T> {
public:
void Compute(const framework::ExecutionContext& ctx) const override {
using DeviceContext = paddle::platform::CPUDeviceContext;
auto* ins = ctx.Input<LoDTensor>("X");
auto* w = ctx.Input<Tensor>("FCWeight");
auto* b = ctx.Input<Tensor>("FCBias");
auto* x = ctx.Input<LoDTensor>("X");
auto* h0 = ctx.Input<Tensor>("H0");
auto* c0 = ctx.Input<Tensor>("C0");
auto* atten_w = ctx.Input<Tensor>("AttentionWeight");
auto* atten_b = ctx.Input<Tensor>("AttentionBias");
auto* atten_scalar = ctx.Input<Tensor>("AttentionScalar");
auto* atten_scalar_bias = ctx.Input<Tensor>("AttentionScalarBias");
auto* lstm_w = ctx.Input<Tensor>("LSTMWeight");
auto* lstm_b = ctx.Input<Tensor>("LSTMBias");
auto* hidden_out = ctx.Output<LoDTensor>("Hidden");
auto* cell_out = ctx.Output<LoDTensor>("Cell");
auto* atted_x = ctx.Output<Tensor>("AttentionedX");
auto* fc_out = ctx.Output<Tensor>("AttentionFCOut");
auto* lstm_x = ctx.Output<Tensor>("LSTMX");
auto* lstm_out = ctx.Output<Tensor>("LSTMOUT");
// some shape should be reshape here since infershape can not get lod info
auto x_lod = x->lod();
const int N = x_lod[0].size() - 1; // batch size
auto x_dims = x->dims(); // T x M
auto w_dims = lstm_w->dims(); // (D+M) x 4D
const int total_T = x_dims[0];
const int M = x_dims[1]; // x frame size
const int D = w_dims[1] / 4; // gate frame size
const int D2 = D * 2;
const int D3 = D * 3;
const int D4 = w_dims[1];
int max_seq_len = x_lod[0][1];
for (int i = 1; i < N; ++i) {
int len = x_lod[0][i + 1] - x_lod[0][i];
max_seq_len = max_seq_len < len ? len : max_seq_len;
}
PADDLE_ENFORCE_EQ(x_lod.size(), 1, "Input(X)'s lod size must be 1.");
PADDLE_ENFORCE_EQ(c0->dims()[0], N, "C0 dims should be %d x %d.", N, D);
fc_out->Resize({max_seq_len, 1});
std::function<void(const int, const T *, T *)> act_gate, act_cell, act_cand;
auto& act_gate_str = ctx.Attr<std::string>("gate_activation");
auto& act_cell_str = ctx.Attr<std::string>("cell_activation");
auto& act_cand_str = ctx.Attr<std::string>("candidate_activation");
auto* out = ctx.Output<LoDTensor>("Out");
auto* fc_out = ctx.Output<Tensor>("FCOUT");
std::function<void(const int, const T*, T*)> fc_act;
auto& fc_act_str = ctx.Attr<std::string>("fc_activation");
if (platform::jit::MayIUse(platform::jit::avx)) {
math::VecActivations<T, platform::jit::avx> act_functor;
act_gate = act_functor(act_gate_str);
act_cell = act_functor(act_cell_str);
act_cand = act_functor(act_cand_str);
fc_act = act_functor(fc_act_str);
} else {
math::VecActivations<T, platform::jit::isa_any> act_functor;
act_gate = act_functor(act_gate_str);
act_cell = act_functor(act_cell_str);
act_cand = act_functor(act_cand_str);
fc_act = act_functor(fc_act_str);
}
const T* x_data = x->data<T>();
const T* h0_data = h0 ? h0->data<T>() : NULL;
const T* c0_data = c0->data<T>();
const T* lstm_w_data = lstm_w->data<T>();
const T* lstm_b_data = lstm_b->data<T>();
const T* atten_w_data = atten_w->data<T>();
const T* atten_b_data = atten_b ? atten_b->data<T>() : NULL;
const T* atten_scalar_data = atten_scalar ? atten_scalar->data<T>() : NULL;
const T* atten_scalar_bias_data =
atten_scalar_bias ? atten_scalar_bias->data<T>() : NULL;
T* hidden_out_data = hidden_out->mutable_data<T>(ctx.GetPlace());
T* cell_out_data = cell_out->mutable_data<T>(ctx.GetPlace());
T* atted_x_data = atted_x->mutable_data<T>(ctx.GetPlace());
PADDLE_ENFORCE_GT(ins.size(), 1, "Input(X)'s size must larger than 1.");
auto* ref_in = ins[0];
auto ref_in_lod = ref_in->lod();
const int N = ref_in_lod[0].size() - 1;
auto ref_in_dims = ref_in->dims(); // T x M0
auto w_dims = w->dims(); // (M0+M1+M2+..) x D
const int total_T = ref_in_dims[0];
const int M0 = ref_in_dims[1];
const int M1 = ins[1]->dims()[1];
const int D = w_dims[1];
const T* ref_in_data =
ref_in->data<T>(); // size should be check at infershape
const T* in1_data = ins[1]->data<T>();
const T* w_data = w->data<T>();
T* out_data = out->mutable_data<T>(ctx.GetPlace());
T* fc_out_data = fc_out->mutable_data<T>(ctx.GetPlace());
T* lstm_x_data = lstm_x->mutable_data<T>(ctx.GetPlace());
T* lstm_out_data = lstm_out->mutable_data<T>(ctx.GetPlace());
// x(TxM) * fc (Mx1) part of atten_wgt(M+D)x1
auto blas = math::GetBlas<DeviceContext, T>(ctx);
math::FCCompute<DeviceContext, T>(blas, total_T, 1, M, x_data, atten_w_data,
atted_x_data, atten_b_data);
const T* cur_atten_x_data = atted_x_data;
const T* cur_x_data = x_data;
const T* prev_cell_data = NULL;
const T* prev_hidden_data = NULL;
T* cur_cell_out_data = cell_out_data;
T* cur_hidden_out_data = hidden_out_data;
math::FCCompute<DeviceContext, T>(blas, total_T, D, M0, ref_in_data, w_data,
out_data, b ? b->data<T>() : NULL);
w_data = w_data + M0 * D;
// first one use write on
blas.MatMul(N, D, M1, in1_data, w_data, fc_out_data);
w_data = w_data + M1 * D;
for (int i = 2; i < ins.size(); ++i) {
// add on
const T* in_data = ins[i]->data<T>();
const int K = ins[i]->dims()[1];
blas.GEMM(CblasNoTrans, CblasNoTrans, N, D, K, static_cast<T>(1), in_data,
K, w_data, D, static_cast<T>(1), fc_out_data, D);
w_data = w_data + K * D;
}
for (int i = 0; i < N; ++i) {
int seq_len = x_lod[0][i + 1] - x_lod[0][i];
prev_cell_data = c0_data + i * D;
prev_hidden_data = h0_data ? h0_data + i * D : NULL;
int seq_len = ref_in_lod[0][i + 1] - ref_in_lod[0][i];
T* src = fc_out_data + i * D;
for (int step = 0; step < seq_len; ++step) {
/// 1. compute attention vector
// 1a. prev_cell(1xD) * fc(D) rest part of atten_wgt
T prev_cell_bias = blas.DOT(D, prev_cell_data, atten_w_data + M);
// 1b. add cell bias and relu
bias_relu<T>(seq_len, cur_atten_x_data, &prev_cell_bias, fc_out_data);
// 1c. fc scalar
if (atten_scalar_data) {
blas.SCAL(seq_len, *atten_scalar_data, fc_out_data);
bias_relu<T>(seq_len, fc_out_data, atten_scalar_bias_data,
fc_out_data);
}
// 1d. softmax
vec_softmax<T>(seq_len, fc_out_data, fc_out_data);
// mul x(seq_len*M) and sum pool
math::FCCompute<DeviceContext, T>(blas, 1, M, seq_len, fc_out_data,
cur_x_data, lstm_x_data);
/// 2. compute LSTM step
// lstm weight : concat[forget , input , output , tilde]
// shape : (D + M) x (4 * D)
// fc inputX(1xM) * weightX(M*(4D)) => 1 x 4D
blas.MatMul(1, D4, M, lstm_x_data, lstm_w_data + D * D4, lstm_out_data);
if (prev_hidden_data) {
blas.GEMM(CblasNoTrans, CblasNoTrans, 1, D4, D, static_cast<T>(1),
prev_hidden_data, D, lstm_w_data, D4, static_cast<T>(1),
lstm_out_data, D4);
}
// since input is 1xM, so can use add bias
blas.VADD(D4, lstm_b_data, lstm_out_data, lstm_out_data);
// gate act: sigmoid
act_gate(D3, lstm_out_data, lstm_out_data);
// candicate act: tanh
act_cand(D, lstm_out_data + D3, lstm_out_data + D3);
// a = forget * prev_cell
blas.VMUL(D, lstm_out_data, prev_cell_data, lstm_out_data);
// b = input * tilde
blas.VMUL(D, lstm_out_data + D, lstm_out_data + D3, lstm_out_data + D);
// cell_out = a + b
blas.VADD(D, lstm_out_data, lstm_out_data + D, cur_cell_out_data);
// state act tanh(cell_out) * output_gate
act_cell(D, cur_cell_out_data, lstm_out_data);
blas.VMUL(D, lstm_out_data, lstm_out_data + D2, cur_hidden_out_data);
prev_hidden_data = cur_hidden_out_data;
prev_cell_data = cur_cell_out_data;
cur_cell_out_data = cur_cell_out_data + D;
cur_hidden_out_data = cur_hidden_out_data + D;
blas.VADD(D, out_data, src, out_data);
out_data = out_data + D;
}
cur_x_data = cur_x_data + seq_len * M;
cur_atten_x_data = cur_atten_x_data + seq_len;
}
fc_act(out_dims[0] * out_dims[1], out_data, out_data);
}
};
......
Markdown is supported
0% 或 .
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 或