apps/FastDeploy
1
0
Fork 0
mirror of https://github.com/PaddlePaddle/FastDeploy.git synced 2025-10-19 23:16:15 +08:00
Code Issues Actions 4 Packages Projects Releases Wiki Activity

[CVCUDA] Vision Processor Python API and Tutorial (#1394)

Browse Source 
* bind success

* bind success fix

* FDMat pybind, ResizeByShort pybind

* FDMat pybind, ResizeByShort pybind, remove initialized_

* override BindProcessorManager::Run in python is available

* PyProcessorManager done

* vision_pybind fix

* manager.py fix

* add tutorials

* remove Apply() bind

* remove Apply() bind and fix

* fix reviewed problem

* fix reviewed problem

* fix reviewed problem readme

* fix reviewed problem readme etc

* apply return outputs

* nits

* update readme

* fix FDMatbatch

* add op pybind: CenterCrop, Pad

* add op overload for pass FDMatBatch

---------

Co-authored-by: Wang Xinyu <shaywxy@gmail.com>
This commit is contained in:
guxukai
2023-03-10 14:42:32 +08:00
committed by GitHub
parent cb7c8a07d4
commit c6480de736
22 changed files with 530 additions and 34 deletions
Download Patch File Download Diff File Expand all files Collapse all files

18
fastdeploy/vision/common/processors/manager_pybind.cc
View File

@@ -14,8 +14,22 @@
#include "fastdeploy/pybind/main.h"
namespace fastdeploy {
namespace vision {
// PyProcessorManager is used for pybind11::init() of ProcessorManager
// Because ProcessorManager have a pure Virtual function Apply()
class FASTDEPLOY_DECL PyProcessorManager : public ProcessorManager {
public:
using ProcessorManager::ProcessorManager;
bool Apply(FDMatBatch* image_batch, std::vector<FDTensor>* outputs) override {
PYBIND11_OVERRIDE_PURE(bool, ProcessorManager, Apply, image_batch, outputs);
}
};
} // namespace vision
void BindProcessorManager(pybind11::module& m) {
pybind11::class_<vision::ProcessorManager>(m, "ProcessorManager")
pybind11::class_<vision::ProcessorManager, vision::PyProcessorManager>(
m, "ProcessorManager")
.def(pybind11::init<>())
.def("run",
[](vision::ProcessorManager& self,
std::vector<pybind11::array>& im_list) {
@@ -34,6 +48,8 @@ void BindProcessorManager(pybind11::module& m) {
}
return outputs;
})
.def("pre_apply", &vision::ProcessorManager::PreApply)
.def("post_apply", &vision::ProcessorManager::PostApply)
.def("use_cuda",
[](vision::ProcessorManager& self, bool enable_cv_cuda = false,
int gpu_id = -1) { self.UseCuda(enable_cv_cuda, gpu_id); });