Skip to content

P
Paddle

PaddlePaddle / Paddle
大约 1 年 前同步成功

通知 2299
Star 20931
Fork 5422
P
Paddle
提交 da9ae643 编写于 作者: X xiaolil1
浏览文件
操作

enable scale reuse

上级 7afea1b6
隐藏空白更改
内联 并排
Showing with 100 addition and 35 deletion
paddle/fluid/operators/conv_mkldnn_op.cc
浏览文件 @ da9ae643
......@@ -15,6 +15,8 @@
#include "paddle/fluid/operators/conv_op.h"
#include "paddle/fluid/platform/mkldnn_helper.h"
#include "paddle/fluid/framework/data_layout_transform.h"
#include <unordered_map>
#include <map>
namespace paddle {
namespace operators {
......@@ -346,36 +348,76 @@ class ConvMKLDNNOpKernel : public paddle::framework::OpKernel<T> {
}
std::vector<int> dst_tz = paddle::framework::vectorize2int(output->dims());
// Get unique name for storing MKLDNN primitives
const std::string key = ConvMKLDNNHandler::GetHash(
src_tz, weights_tz, strides, paddings, dilations, groups,
ctx.op().Output("Output"));
const std::string key_conv_pd = key + "@conv_pd";
static std::unordered_map<std::string, std::vector<float>> scale_map;
//scale_map.insert({key_conv_pd,{1.0f}});
//scale_map[key_conv_pd]={0.1f};
bool scale_reuse = false;
auto scale_in_key = key + "@scale_in";
auto scale_weights_key = key + "@scale_weights";
auto scale_out_key = key + "@scale_out";
auto output_shift_scale_key = key + "@output_shift_scale";
auto sum_scale_key = key + "@sum_scale";
auto scale_in_eltwise_key = key + "@scale_in_eltwise";
std::vector<float> scale_in_data;
std::vector<float> scale_out_data;
std::vector<float> scale_weights_data;
std::vector<float> scale_in_eltwise_data;
std::vector<float> output_shift_scale;
float sum_scale = 1.0f;
std::vector<float> sum_scale = {1.0f};
std::vector<float> none_scale = {0};
if (is_INT8 && GetScaleMap(scale_map, scale_in_key) == none_scale){
scale_reuse = true;
}
//std::cout<<"scale_reuse = "<<scale_reuse<<std::endl;
if(is_INT8){
int count = is_multi_channel? (g>1? weights_tz[1]*weights_tz[0] : weights_tz[0]) : 1;
float scale_in_data = *(scale_in->data<float>());
std::vector<float> scale_weights_data(count);
for(int i=0; i<count; i++){
scale_weights_data[i] =*(scale_weights->data<float>() + i);
}
float scale_out_data = *(scale_out->data<float>());
output_shift_scale.resize(count);
for(int i=0; i<count; i++){
if(scale_weights_data[i] == 0.0)
output_shift_scale[i] = scale_out_data;
else
output_shift_scale[i] = scale_out_data / (scale_in_data * scale_weights_data[i]);
}
if(fuse_residual_conn){
float scale_in_eltwise_data = *(scale_in_eltwise->data<float>());
sum_scale = scale_out_data / scale_in_eltwise_data;
if(scale_reuse){
//std::cout<<"load scale!!!!!!!!"<<std::endl;
int count = is_multi_channel? (g>1? weights_tz[1]*weights_tz[0] : weights_tz[0]) : 1;
scale_in_data = {*(scale_in->data<float>())};
scale_weights_data.resize(count);
for(int i=0; i<count; i++){
scale_weights_data[i] =*(scale_weights->data<float>() + i);
}
scale_out_data = {*(scale_out->data<float>())};
output_shift_scale.resize(count);
for(int i=0; i<count; i++){
if(scale_weights_data[i] == 0.0)
output_shift_scale[i] = scale_out_data[0];
else
output_shift_scale[i] = scale_out_data[0] / (scale_in_data[0] * scale_weights_data[i]);
}
if(fuse_residual_conn){
scale_in_eltwise_data = {*(scale_in_eltwise->data<float>())};
sum_scale[0] = scale_out_data[0] / scale_in_eltwise_data[0];
SetScaleMap(scale_map, scale_in_eltwise_key, scale_in_eltwise_data);
}
//scale reuse
SetScaleMap(scale_map, scale_in_key, scale_in_data);
SetScaleMap(scale_map, scale_weights_key, scale_weights_data);
SetScaleMap(scale_map, scale_out_key, scale_out_data);
SetScaleMap(scale_map, output_shift_scale_key, output_shift_scale);
SetScaleMap(scale_map, sum_scale_key, sum_scale);
} else{
scale_in_data = GetScaleMap(scale_map, scale_in_key);
scale_out_data = GetScaleMap(scale_map, scale_out_key);
scale_weights_data = GetScaleMap(scale_map, scale_weights_key);
if(fuse_residual_conn){
scale_in_eltwise_data = GetScaleMap(scale_map, scale_in_eltwise_key);
}
output_shift_scale = GetScaleMap(scale_map, output_shift_scale_key);
sum_scale = GetScaleMap(scale_map, sum_scale_key);
//printf("pause!!!");
}
}
// Get unique name for storing MKLDNN primitives
const std::string key = ConvMKLDNNHandler::GetHash(
src_tz, weights_tz, strides, paddings, dilations, groups,
ctx.op().Output("Output"));
const std::string key_conv_pd = key + "@conv_pd";
std::vector<primitive> pipeline;
auto user_src_md = platform::MKLDNNMemDesc(
......@@ -431,7 +473,7 @@ class ConvMKLDNNOpKernel : public paddle::framework::OpKernel<T> {
conv_pd = ConvFwdPrimitiveDesc(src_md, weights_md, bias_md, dst_md,
strides, paddings, mkldnn_engine,
fuse_relu, fuse_residual_conn,
output_shift_scale, sum_scale);
output_shift_scale, sum_scale[0]);
} else{
conv_pd = ConvFwdPrimitiveDesc(src_md, weights_md, bias_md, dst_md,
strides, paddings, mkldnn_engine,
......@@ -442,7 +484,7 @@ class ConvMKLDNNOpKernel : public paddle::framework::OpKernel<T> {
conv_pd =
ConvFwdPrimitiveDesc(src_md, weights_md, dst_md, strides, paddings,
mkldnn_engine, fuse_relu, fuse_residual_conn,
output_shift_scale, sum_scale);
output_shift_scale, sum_scale[0]);
} else{
conv_pd =
ConvFwdPrimitiveDesc(src_md, weights_md, dst_md, strides, paddings,
......@@ -466,11 +508,6 @@ class ConvMKLDNNOpKernel : public paddle::framework::OpKernel<T> {
std::shared_ptr<mkldnn::memory> weights_memory_p;
if(is_INT8){
int mask_reorder = is_multi_channel? ((g!= 1) ? (1<<1)+(1<<0) : 1<<0) : 0;
int count = is_multi_channel? (g>1? weights_tz[1]*weights_tz[0] : weights_tz[0]) : 1;
std::vector<float> scale_weights_data(count);
for(int i=0; i<count; i++){
scale_weights_data[i] = *(scale_weights->data<float>() + i);
}
weights_memory_p = handler.AcquireWeightsMemoryFromPrimitive(
user_weights_memory_p, pipeline, is_test, is_INT8, scale_weights_data, mask_reorder);
} else{
......@@ -536,6 +573,8 @@ class ConvMKLDNNOpKernel : public paddle::framework::OpKernel<T> {
// create convolution op primitive
std::shared_ptr<mkldnn::convolution_forward> conv_p;
std::vector<float> scale_bias_data;
auto scale_bias_key = key + "@scale_bias";
if (bias) {
const float* bias_data = bias->data<float>();
auto user_bias_md = platform::MKLDNNMemDesc(
......@@ -545,10 +584,15 @@ class ConvMKLDNNOpKernel : public paddle::framework::OpKernel<T> {
std::shared_ptr<mkldnn::memory> bias_memory_p;
if(is_INT8){
int mask_reorder = is_multi_channel? 1<<0 : 1;
int count = is_multi_channel? (g>1? weights_tz[1]*weights_tz[0] : weights_tz[0]) : 1;
std::vector<float> scale_bias_data(count);
for(int i=0; i<count; i++){
scale_bias_data[i] = (*scale_in->data<float>()) * (*(scale_weights->data<float>() + i));
if(scale_reuse){
int count = is_multi_channel? (g>1? weights_tz[1]*weights_tz[0] : weights_tz[0]) : 1;
scale_bias_data.resize(count);
for(int i=0; i<count; i++){
scale_bias_data[i] = scale_in_data[0] * scale_weights_data[i];
}
SetScaleMap(scale_map, scale_bias_key, scale_bias_data);
} else{
scale_bias_data = GetScaleMap(scale_map, scale_bias_key);
}
bias_memory_p =
handler.AcquireBiasMemoryFromPrimitive(user_bias_memory_p, pipeline, is_test, is_INT8, scale_bias_data, mask_reorder);
......@@ -577,6 +621,27 @@ class ConvMKLDNNOpKernel : public paddle::framework::OpKernel<T> {
}
private:
void SetScaleMap(std::unordered_map<std::string, std::vector<float>> &scale_map,
const std::string& name, std::vector<float> scale_data) const {
auto it = scale_map.find(name);
if (it == scale_map.end()) {
scale_map[name] = scale_data; // create new blob
} else {
(*it).second = scale_data; // set data to existing blob
}
return;
}
std::vector<float> GetScaleMap(std::unordered_map<std::string, std::vector<float>> &scale_map,
const std::string& name) const {
auto it = scale_map.find(name);
if (it != scale_map.end()) {
return (*it).second;
}
return {0};
}
mkldnn::primitive_attr CreatePostOps(bool fuse_relu, bool fuse_residual_conn,
const std::vector<float> output_shift_scale, float sum_scale) const {
mkldnn::primitive_attr conv_attr;
......
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