Skip to content
体验新版
项目
组织
正在加载...
登录
切换导航
打开侧边栏
csdn_franckjun
Paddle
提交
a9ed4277
P
Paddle
项目概览
csdn_franckjun
/
Paddle
与 Fork 源项目一致
Fork自
PaddlePaddle / Paddle
通知
1
Star
1
Fork
0
代码
文件
提交
分支
Tags
贡献者
分支图
Diff
Issue
1
列表
看板
标记
里程碑
合并请求
0
Wiki
0
Wiki
分析
仓库
DevOps
项目成员
Pages
P
Paddle
项目概览
项目概览
详情
发布
仓库
仓库
文件
提交
分支
标签
贡献者
分支图
比较
Issue
1
Issue
1
列表
看板
标记
里程碑
合并请求
0
合并请求
0
Pages
分析
分析
仓库分析
DevOps
Wiki
0
Wiki
成员
成员
收起侧边栏
关闭侧边栏
动态
分支图
创建新Issue
提交
Issue看板
体验新版 GitCode,发现更多精彩内容 >>
提交
a9ed4277
编写于
3月 07, 2019
作者:
N
nhzlx
浏览文件
操作
浏览文件
下载
电子邮件补丁
差异文件
cant not pass ci
add if use static engine for trt test=develop
上级
898d7d8b
变更
8
隐藏空白更改
内联
并排
Showing
8 changed file
with
21 addition
and
5 deletion
+21
-5
paddle/fluid/inference/analysis/argument.h
paddle/fluid/inference/analysis/argument.h
+6
-0
paddle/fluid/inference/analysis/ir_pass_manager.cc
paddle/fluid/inference/analysis/ir_pass_manager.cc
+2
-0
paddle/fluid/inference/analysis/ir_passes/tensorrt_subgraph_pass.cc
...id/inference/analysis/ir_passes/tensorrt_subgraph_pass.cc
+2
-1
paddle/fluid/inference/api/analysis_config.cc
paddle/fluid/inference/api/analysis_config.cc
+3
-1
paddle/fluid/inference/api/analysis_predictor.cc
paddle/fluid/inference/api/analysis_predictor.cc
+1
-0
paddle/fluid/inference/api/paddle_analysis_config.h
paddle/fluid/inference/api/paddle_analysis_config.h
+3
-1
paddle/fluid/inference/tests/api/trt_models_tester.cc
paddle/fluid/inference/tests/api/trt_models_tester.cc
+2
-1
paddle/fluid/pybind/inference_api.cc
paddle/fluid/pybind/inference_api.cc
+2
-1
未找到文件。
paddle/fluid/inference/analysis/argument.h
浏览文件 @
a9ed4277
...
...
@@ -23,8 +23,12 @@
#pragma once
#include <memory>
#include <string>
#include <unordered_map>
#include <unordered_set>
#include <vector>
#include "paddle/fluid/framework/ir/graph.h"
#include "paddle/fluid/framework/program_desc.h"
#include "paddle/fluid/framework/scope.h"
...
...
@@ -133,6 +137,8 @@ struct Argument {
DECL_ARGUMENT_FIELD
(
tensorrt_min_subgraph_size
,
TensorRtMinSubgraphSize
,
int
);
DECL_ARGUMENT_FIELD
(
tensorrt_precision_mode
,
TensorRtPrecisionMode
,
AnalysisConfig
::
Precision
);
DECL_ARGUMENT_FIELD
(
tensorrt_use_static_engine
,
TensorRtUseStaticEngine
,
bool
);
// Memory optimized related.
DECL_ARGUMENT_FIELD
(
enable_memory_optim
,
EnableMemoryOptim
,
bool
);
...
...
paddle/fluid/inference/analysis/ir_pass_manager.cc
浏览文件 @
a9ed4277
...
...
@@ -82,6 +82,8 @@ void IRPassManager::CreatePasses(Argument *argument,
"model_opt_cache_dir"
,
new
std
::
string
(
GetOrCreateModelOptCacheDir
(
model_opt_cache_dir
)));
pass
->
Set
(
"gpu_device_id"
,
new
int
(
argument
->
gpu_device_id
()));
pass
->
Set
(
"use_static_engine"
,
new
bool
(
argument
->
tensorrt_use_static_engine
()));
}
pre_pass
=
pass_name
;
...
...
paddle/fluid/inference/analysis/ir_passes/tensorrt_subgraph_pass.cc
浏览文件 @
a9ed4277
...
...
@@ -226,10 +226,11 @@ void TensorRtSubgraphPass::CreateTensorRTOp(
calibrator
.
reset
(
new
tensorrt
::
TRTInt8Calibrator
(
calibration_data
));
}
bool
use_static_engine
=
Get
<
bool
>
(
"use_static_engine"
);
// When in int8 mode and calibration_mode, the program just produce the
// calibration table data.
bool
calibration_mode
=
(
enable_int8
&&
calibration_data
.
size
()
==
0
);
if
(
!
calibration_mode
)
{
if
(
!
calibration_mode
&&
use_static_engine
)
{
std
::
copy
(
params
.
begin
(),
params
.
end
(),
std
::
back_inserter
(
*
repetitive_params
));
std
::
string
trt_engine_serialized_data
=
GetTrtEngineSerializedData
(
...
...
paddle/fluid/inference/api/analysis_config.cc
浏览文件 @
a9ed4277
...
...
@@ -103,6 +103,7 @@ AnalysisConfig::AnalysisConfig(const AnalysisConfig &other) {
CP_MEMBER
(
tensorrt_max_batchsize_
);
CP_MEMBER
(
tensorrt_min_subgraph_size_
);
CP_MEMBER
(
tensorrt_precision_mode_
);
CP_MEMBER
(
trt_use_static_engine_
);
// MKLDNN related.
CP_MEMBER
(
use_mkldnn_
);
CP_MEMBER
(
mkldnn_enabled_op_types_
);
...
...
@@ -144,7 +145,7 @@ void AnalysisConfig::EnableMKLDNN() {
void
AnalysisConfig
::
EnableTensorRtEngine
(
int
workspace_size
,
int
max_batch_size
,
int
min_subgraph_size
,
AnalysisConfig
::
Precision
precision_mode
)
{
AnalysisConfig
::
Precision
precision_mode
,
bool
use_static
)
{
#ifdef PADDLE_WITH_CUDA
if
(
!
use_gpu
())
{
LOG
(
ERROR
)
<<
"To use TensorRT engine, please call EnableGpu() first"
;
...
...
@@ -156,6 +157,7 @@ void AnalysisConfig::EnableTensorRtEngine(
tensorrt_max_batchsize_
=
max_batch_size
;
tensorrt_min_subgraph_size_
=
min_subgraph_size
;
tensorrt_precision_mode_
=
precision_mode
;
trt_use_static_engine_
=
use_static
;
Update
();
#else
...
...
paddle/fluid/inference/api/analysis_predictor.cc
浏览文件 @
a9ed4277
...
...
@@ -362,6 +362,7 @@ void AnalysisPredictor::OptimizeInferenceProgram() {
argument_
.
SetTensorRtMaxBatchSize
(
config_
.
tensorrt_max_batchsize_
);
argument_
.
SetTensorRtMinSubgraphSize
(
config_
.
tensorrt_min_subgraph_size_
);
argument_
.
SetTensorRtPrecisionMode
(
config_
.
tensorrt_precision_mode_
);
argument_
.
SetTensorRtUseStaticEngine
(
config_
.
trt_use_static_engine_
);
}
if
(
config_
.
use_mkldnn_
)
{
...
...
paddle/fluid/inference/api/paddle_analysis_config.h
浏览文件 @
a9ed4277
...
...
@@ -135,7 +135,8 @@ struct AnalysisConfig {
*/
void
EnableTensorRtEngine
(
int
workspace_size
=
1
<<
20
,
int
max_batch_size
=
1
,
int
min_subgraph_size
=
3
,
Precision
precision
=
Precision
::
kFloat32
);
Precision
precision
=
Precision
::
kFloat32
,
bool
use_static
=
true
);
/** A boolean state telling whether the TensorRT engine is used.
*/
bool
tensorrt_engine_enabled
()
const
{
return
use_tensorrt_
;
}
...
...
@@ -233,6 +234,7 @@ struct AnalysisConfig {
// subgraph, 3 as default value.
int
tensorrt_min_subgraph_size_
{
3
};
Precision
tensorrt_precision_mode_
;
bool
trt_use_static_engine_
;
// memory reuse related.
bool
enable_memory_optim_
{
false
};
...
...
paddle/fluid/inference/tests/api/trt_models_tester.cc
浏览文件 @
a9ed4277
...
...
@@ -54,7 +54,8 @@ void SetConfig<AnalysisConfig>(AnalysisConfig* config, std::string model_dir,
if
(
use_gpu
)
{
config
->
EnableUseGpu
(
100
,
0
);
if
(
use_tensorrt
)
{
config
->
EnableTensorRtEngine
(
1
<<
10
,
batch_size
);
config
->
EnableTensorRtEngine
(
1
<<
10
,
batch_size
,
3
,
AnalysisConfig
::
Precision
::
kFloat32
,
false
);
config
->
pass_builder
()
->
DeletePass
(
"conv_bn_fuse_pass"
);
config
->
pass_builder
()
->
DeletePass
(
"fc_fuse_pass"
);
config
->
pass_builder
()
->
TurnOnDebug
();
...
...
paddle/fluid/pybind/inference_api.cc
浏览文件 @
a9ed4277
...
...
@@ -221,7 +221,8 @@ void BindAnalysisConfig(py::module *m) {
.
def
(
"enable_tensorrt_engine"
,
&
AnalysisConfig
::
EnableTensorRtEngine
,
py
::
arg
(
"workspace_size"
)
=
1
<<
20
,
py
::
arg
(
"max_batch_size"
)
=
1
,
py
::
arg
(
"min_subgraph_size"
)
=
3
,
py
::
arg
(
"precision_mode"
)
=
AnalysisConfig
::
Precision
::
kFloat32
)
py
::
arg
(
"precision_mode"
)
=
AnalysisConfig
::
Precision
::
kFloat32
,
py
::
arg
(
"use_static"
)
=
true
)
.
def
(
"tensorrt_engine_enabled"
,
&
AnalysisConfig
::
tensorrt_engine_enabled
)
.
def
(
"switch_ir_debug"
,
&
AnalysisConfig
::
SwitchIrDebug
,
py
::
arg
(
"x"
)
=
true
)
...
...
编辑
预览
Markdown
is supported
0%
请重试
或
添加新附件
.
添加附件
取消
You are about to add
0
people
to the discussion. Proceed with caution.
先完成此消息的编辑!
取消
想要评论请
注册
或
登录