apps/FastDeploy
1
0
Fork 0
mirror of https://github.com/PaddlePaddle/FastDeploy.git synced 2025-10-05 16:48:03 +08:00
Code Issues Actions 2 Packages Projects Releases Wiki Activity

[Benchmark]Add trt cache option (#1653)

Browse Source 
* add GPL lisence

* add GPL-3.0 lisence

* add GPL-3.0 lisence

* add GPL-3.0 lisence

* support yolov8

* add pybind for yolov8

* add yolov8 readme

* add cpp benchmark

* add cpu and gpu mem

* public part split

* add runtime mode

* fixed bugs

* add cpu_thread_nums

* deal with comments

* deal with comments

* deal with comments

* rm useless code

* add FASTDEPLOY_DECL

* add FASTDEPLOY_DECL

* fixed for windows

* mv rss to pss

* mv rss to pss

* Update utils.cc

* use thread to collect mem

* Add ResourceUsageMonitor

* rm useless code

* fixed bug

* fixed typo

* update ResourceUsageMonitor

* fixed bug

* fixed bug

* add note for ResourceUsageMonitor

* deal with comments

* add macros

* deal with comments

* deal with comments

* deal with comments

* re-lint

* rm pmap and use mem api

* rm pmap and use mem api

* add mem api

* Add PrintBenchmarkInfo func

* Add PrintBenchmarkInfo func

* Add PrintBenchmarkInfo func

* deal with comments

* fixed enable_paddle_to_trt

* add log for paddle_trt

* support ppcls benchmark

* use new trt option api

* update benchmark info

* simplify benchmark.cc

* simplify benchmark.cc

* deal with comments

* Add ppseg && ppocr benchmark

* add OCR rec img

* add ocr benchmark

* fixed trt shape

* add trt shape

* resolve conflict

* add ENABLE_BENCHMARK define

* Add ClassifyDiff

* Add Resize for ClassifyResult

* deal with comments

* add convert info script

* resolve conflict

* Add SaveBenchmarkResult func

* fixed bug

* fixed bug

* fixed bug

* add config.txt for option

* fixed bug

* fixed bug

* fixed bug

* add benchmark.sh

* mv thread_nums from 8 to 1

* deal with comments

* deal with comments

* fixed readme

* deal with comments

* add all platform shell

* Update config.arm.txt

* Update config.gpu.txt

* Update config.x86.txt

* fixed printinfo bug

* rm proxy

* add more model support

* all backend config.txt

* deal with comments

* Add MattingDiff compare

* fixed predict bug

* adjust warmup/repeat times

* add e2e/mem configs

* fixed typo

* open collect_mem

* fixed typo

* add trt cache option

* fixed bug

* fixed repeat times

---------

Co-authored-by: DefTruth <31974251+DefTruth@users.noreply.github.com>
This commit is contained in:
WJJ1995
2023-03-20 18:08:28 +08:00
committed by GitHub
parent d0de379d9a
commit d522ef28b4
38 changed files with 36 additions and 29 deletions
Download Patch File Download Diff File Expand all files Collapse all files

2
benchmark/cpp/benchmark_gpu_trt.sh
View File

@@ -44,4 +44,4 @@ fi
./benchmark_ppocr_cls --model ch_ppocr_mobile_v2.0_cls_infer --image rec_img.jpg --config_path $CONFIG_PATH --trt_shape 1,3,48,10:4,3,48,320:8,3,48,1024 ./benchmark_ppocr_cls --model ch_ppocr_mobile_v2.0_cls_infer --image rec_img.jpg --config_path $CONFIG_PATH --trt_shape 1,3,48,10:4,3,48,320:8,3,48,1024
./benchmark_ppocr_rec --model ch_PP-OCRv3_rec_infer --image rec_img.jpg --rec_label_file ppocr_keys_v1.txt --config_path $CONFIG_PATH --trt_shape 1,3,48,10:4,3,48,320:8,3,48,2304 ./benchmark_ppocr_rec --model ch_PP-OCRv3_rec_infer --image rec_img.jpg --rec_label_file ppocr_keys_v1.txt --config_path $CONFIG_PATH --trt_shape 1,3,48,10:4,3,48,320:8,3,48,2304
./benchmark_ppocr_det --model ch_PP-OCRv2_det_infer --image 12.jpg --config_path $CONFIG_PATH --trt_shape 1,3,64,64:1,3,640,640:1,3,960,960 ./benchmark_ppocr_det --model ch_PP-OCRv2_det_infer --image 12.jpg --config_path $CONFIG_PATH --trt_shape 1,3,64,64:1,3,640,640:1,3,960,960
./benchmark_ppocr_rec --model ch_PP-OCRv2_rec_infer --image rec_img.jpg --rec_label_file ppocr_keys_v1.txt --config_path $CONFIG_PATH --trt_shape 1,3,48,10:4,3,48,320:8,3,48,2304 ./benchmark_ppocr_rec --model ch_PP-OCRv2_rec_infer --image rec_img.jpg --rec_label_file ppocr_keys_v1.txt --config_path $CONFIG_PATH --trt_shape 1,3,32,10:4,3,32,320:8,3,32,2304

0
benchmark/cpp/benchmark_picodet.cc Normal file → Executable file
View File

0
benchmark/cpp/benchmark_ppmatting.cc Normal file → Executable file
View File

0
benchmark/cpp/benchmark_ppocr_cls.cc Normal file → Executable file
View File

0
benchmark/cpp/benchmark_ppocr_det.cc Normal file → Executable file
View File

5
benchmark/cpp/benchmark_ppocr_rec.cc Normal file → Executable file
View File

@@ -55,6 +55,11 @@ int main(int argc, char* argv[]) {
} }
auto model_ppocr_rec = vision::ocr::Recognizer( auto model_ppocr_rec = vision::ocr::Recognizer(
model_file, params_file, FLAGS_rec_label_file, option, model_format); model_file, params_file, FLAGS_rec_label_file, option, model_format);
std::vector<std::string> model_names;
fastdeploy::benchmark::Split(FLAGS_model, model_names, sep);
if (model_names[model_names.size() - 1] == "ch_PP-OCRv2_rec_infer") {
model_ppocr_rec.GetPreprocessor().SetRecImageShape({3, 32, 320});
}
std::string text; std::string text;
float rec_score; float rec_score;
if (config_info["precision_compare"] == "true") { if (config_info["precision_compare"] == "true") {

0
benchmark/cpp/benchmark_ppseg.cc Normal file → Executable file
View File

0
benchmark/cpp/benchmark_ppyoloe.cc Normal file → Executable file
View File

0
benchmark/cpp/benchmark_ppyolov5.cc Normal file → Executable file
View File

0
benchmark/cpp/benchmark_ppyolov6.cc Normal file → Executable file
View File

0
benchmark/cpp/benchmark_ppyolov7.cc Normal file → Executable file
View File

0
benchmark/cpp/benchmark_ppyolov8.cc Normal file → Executable file
View File

0
benchmark/cpp/benchmark_ppyolox.cc Normal file → Executable file
View File

4
benchmark/cpp/config/config.gpu.ort.fp32.e2e.mem.txt
View File

@@ -1,8 +1,8 @@
device: gpu device: gpu
device_id: 3 device_id: 3
cpu_thread_nums: 1 cpu_thread_nums: 1
warmup: 10 warmup: 20
repeat: 50 repeat: 100
backend: ort backend: ort
profile_mode: end2end profile_mode: end2end
include_h2d_d2h: false include_h2d_d2h: false

4
benchmark/cpp/config/config.gpu.paddle.fp32.e2e.mem.txt
View File

@@ -1,8 +1,8 @@
device: gpu device: gpu
device_id: 3 device_id: 3
cpu_thread_nums: 1 cpu_thread_nums: 1
warmup: 10 warmup: 20
repeat: 50 repeat: 100
backend: paddle backend: paddle
profile_mode: end2end profile_mode: end2end
include_h2d_d2h: false include_h2d_d2h: false

4
benchmark/cpp/config/config.gpu.paddle_trt.fp16.e2e.mem.txt
View File

@@ -1,8 +1,8 @@
device: gpu device: gpu
device_id: 3 device_id: 3
cpu_thread_nums: 1 cpu_thread_nums: 1
warmup: 10 warmup: 20
repeat: 50 repeat: 100
backend: paddle_trt backend: paddle_trt
profile_mode: end2end profile_mode: end2end
include_h2d_d2h: false include_h2d_d2h: false

4
benchmark/cpp/config/config.gpu.paddle_trt.fp16.e2e.txt
View File

@@ -1,8 +1,8 @@
device: gpu device: gpu
device_id: 3 device_id: 3
cpu_thread_nums: 1 cpu_thread_nums: 1
warmup: 20 warmup: 200
repeat: 100 repeat: 1000
backend: paddle_trt backend: paddle_trt
profile_mode: end2end profile_mode: end2end
include_h2d_d2h: false include_h2d_d2h: false

4
benchmark/cpp/config/config.gpu.paddle_trt.fp16.txt
View File

@@ -1,8 +1,8 @@
device: gpu device: gpu
device_id: 3 device_id: 3
cpu_thread_nums: 1 cpu_thread_nums: 1
warmup: 20 warmup: 200
repeat: 100 repeat: 1000
backend: paddle_trt backend: paddle_trt
profile_mode: runtime profile_mode: runtime
include_h2d_d2h: false include_h2d_d2h: false

4
benchmark/cpp/config/config.gpu.paddle_trt.fp32.e2e.mem.txt
View File

@@ -1,8 +1,8 @@
device: gpu device: gpu
device_id: 3 device_id: 3
cpu_thread_nums: 1 cpu_thread_nums: 1
warmup: 10 warmup: 20
repeat: 50 repeat: 100
backend: paddle_trt backend: paddle_trt
profile_mode: end2end profile_mode: end2end
include_h2d_d2h: false include_h2d_d2h: false

4
benchmark/cpp/config/config.gpu.paddle_trt.fp32.e2e.txt
View File

@@ -1,8 +1,8 @@
device: gpu device: gpu
device_id: 3 device_id: 3
cpu_thread_nums: 1 cpu_thread_nums: 1
warmup: 20 warmup: 200
repeat: 100 repeat: 1000
backend: paddle_trt backend: paddle_trt
profile_mode: end2end profile_mode: end2end
include_h2d_d2h: false include_h2d_d2h: false

4
benchmark/cpp/config/config.gpu.paddle_trt.fp32.txt
View File

@@ -1,8 +1,8 @@
device: gpu device: gpu
device_id: 3 device_id: 3
cpu_thread_nums: 1 cpu_thread_nums: 1
warmup: 20 warmup: 200
repeat: 100 repeat: 1000
backend: paddle_trt backend: paddle_trt
profile_mode: runtime profile_mode: runtime
include_h2d_d2h: false include_h2d_d2h: false

4
benchmark/cpp/config/config.gpu.trt.fp16.e2e.mem.txt
View File

@@ -1,8 +1,8 @@
device: gpu device: gpu
device_id: 3 device_id: 3
cpu_thread_nums: 1 cpu_thread_nums: 1
warmup: 10 warmup: 20
repeat: 50 repeat: 100
backend: trt backend: trt
profile_mode: end2end profile_mode: end2end
include_h2d_d2h: false include_h2d_d2h: false

4
benchmark/cpp/config/config.gpu.trt.fp16.e2e.txt
View File

@@ -1,8 +1,8 @@
device: gpu device: gpu
device_id: 3 device_id: 3
cpu_thread_nums: 1 cpu_thread_nums: 1
warmup: 20 warmup: 200
repeat: 100 repeat: 1000
backend: trt backend: trt
profile_mode: end2end profile_mode: end2end
include_h2d_d2h: false include_h2d_d2h: false

4
benchmark/cpp/config/config.gpu.trt.fp16.txt
View File

@@ -1,8 +1,8 @@
device: gpu device: gpu
device_id: 3 device_id: 3
cpu_thread_nums: 1 cpu_thread_nums: 1
warmup: 20 warmup: 200
repeat: 100 repeat: 1000
backend: trt backend: trt
profile_mode: runtime profile_mode: runtime
include_h2d_d2h: false include_h2d_d2h: false

4
benchmark/cpp/config/config.gpu.trt.fp32.e2e.mem.txt
View File

@@ -1,8 +1,8 @@
device: gpu device: gpu
device_id: 3 device_id: 3
cpu_thread_nums: 1 cpu_thread_nums: 1
warmup: 10 warmup: 20
repeat: 50 repeat: 100
backend: trt backend: trt
profile_mode: end2end profile_mode: end2end
include_h2d_d2h: false include_h2d_d2h: false

4
benchmark/cpp/config/config.gpu.trt.fp32.e2e.txt
View File

@@ -1,8 +1,8 @@
device: gpu device: gpu
device_id: 3 device_id: 3
cpu_thread_nums: 1 cpu_thread_nums: 1
warmup: 20 warmup: 200
repeat: 100 repeat: 1000
backend: trt backend: trt
profile_mode: end2end profile_mode: end2end
include_h2d_d2h: false include_h2d_d2h: false

4
benchmark/cpp/config/config.gpu.trt.fp32.txt
View File

@@ -1,8 +1,8 @@
device: gpu device: gpu
device_id: 3 device_id: 3
cpu_thread_nums: 1 cpu_thread_nums: 1
warmup: 20 warmup: 200
repeat: 100 repeat: 1000
backend: trt backend: trt
profile_mode: runtime profile_mode: runtime
include_h2d_d2h: false include_h2d_d2h: false

2
benchmark/cpp/option.h
View File

@@ -43,6 +43,8 @@ static bool CreateRuntimeOption(fastdeploy::RuntimeOption* option,
option->UsePaddleInferBackend(); option->UsePaddleInferBackend();
} else if (config_info["backend"] == "trt" || } else if (config_info["backend"] == "trt" ||
config_info["backend"] == "paddle_trt") { config_info["backend"] == "paddle_trt") {
option->trt_option.serialize_file = FLAGS_model +
sep + "trt_serialized.trt";
option->UseTrtBackend(); option->UseTrtBackend();
if (config_info["backend"] == "paddle_trt") { if (config_info["backend"] == "paddle_trt") {
option->UsePaddleInferBackend(); option->UsePaddleInferBackend();

0
benchmark/python/README.md Normal file → Executable file
View File

0
benchmark/python/benchmark_ernie_seq_cls.py Normal file → Executable file
View File

0
benchmark/python/convert_info.py Normal file → Executable file
View File

0
benchmark/python/requirements.txt Normal file → Executable file
View File

0
benchmark/python/run_benchmark_ernie_seq_cls.sh Normal file → Executable file
View File

0
benchmark/python/run_benchmark_ppcls.sh Normal file → Executable file
View File

0
benchmark/python/run_benchmark_ppdet.sh Normal file → Executable file
View File

0
benchmark/python/run_benchmark_ppocr.sh Normal file → Executable file
View File

0
benchmark/python/run_benchmark_ppseg.sh Normal file → Executable file
View File

0
benchmark/python/run_benchmark_uie.sh Normal file → Executable file
View File