apps/FastDeploy
1
0
Fork 0
mirror of https://github.com/PaddlePaddle/FastDeploy.git synced 2025-10-05 08:37:06 +08:00
Code Issues Actions 2 Packages Projects Releases Wiki Activity

Add paddle quantize model support for ORT, TRT and MKLDNN deploy backend (#257)

Browse Source 
* add quantize model support for trt and paddle

* fix bugs

* fix

* update paddle2onnx version

* update version

* add quantize test

Co-authored-by: Jason <jiangjiajun@baidu.com>
This commit is contained in:
yeliang2258
2022-10-09 20:00:05 +08:00
committed by GitHub
parent ff5e798b7f
commit 2a68a23baf
10 changed files with 187 additions and 5 deletions
Download Patch File Download Diff File Expand all files Collapse all files

7
fastdeploy/runtime.cc Normal file → Executable file
View File

@@ -198,6 +198,13 @@ void RuntimeOption::SetCpuThreadNum(int thread_num) {
cpu_thread_num = thread_num;
}
void RuntimeOption::SetOrtGraphOptLevel(int level) {
std::vector<int> supported_level{-1, 0, 1, 2};
auto valid_level = std::find(supported_level.begin(), supported_level.end(), level) != supported_level.end();
FDASSERT(valid_level, "The level must be -1, 0, 1, 2.");
ort_graph_opt_level = level;
}
// use paddle inference backend
void RuntimeOption::UsePaddleBackend() {
#ifdef ENABLE_PADDLE_BACKEND