mmdeploy/csrc/core/graph.h
lvhan028 36124f6205
Merge sdk (#251)
* check in cmake

* move backend_ops to csrc/backend_ops

* check in preprocess, model, some codebase and their c-apis

* check in CMakeLists.txt

* check in parts of test_csrc

* commit everything else

* add readme

* update core's BUILD_INTERFACE directory

* skip codespell on third_party

* update trt_net and ort_net's CMakeLists

* ignore clion's build directory

* check in pybind11

* add onnx.proto. Remove MMDeploy's dependency on ncnn's source code

* export MMDeployTargets only when MMDEPLOY_BUILD_SDK is ON

* remove useless message

* target include directory is wrong

* change target name from mmdeploy_ppl_net to mmdeploy_pplnn_net

* skip install directory

* update project's cmake

* remove useless code

* set CMAKE_BUILD_TYPE to Release by force if it isn't set by user

* update custom ops CMakeLists

* pass object target's source lists

* fix lint end-of-file

* fix lint: trailing whitespace

* fix codespell hook

* remove bicubic_interpolate to csrc/backend_ops/

* set MMDEPLOY_BUILD_SDK OFF

* change custom ops build command

* add spdlog installation command

* update docs on how to checkout pybind11

* move bicubic_interpolate to backend_ops/tensorrt directory

* remove useless code

* correct cmake

* fix typo

* fix typo

* fix install directory

* correct sdk's readme

* set cub dir when cuda version < 11.0

* change directory where clang-format will apply to

* fix build command

* add .clang-format

* change clang-format style from google to file

* reformat csrc/backend_ops

* format sdk's code

* turn off clang-format for some files

* add -Xcompiler=-fno-gnu-unique

* fix trt topk initialize

* check in config for sdk demo

* update cmake script and csrc's readme

* correct config's path

* add cuda include directory, otherwise compile failed in case of tensorrt8.2

* clang-format onnx2ncnn.cpp

Co-authored-by: zhangli <lzhang329@gmail.com>
Co-authored-by: grimoire <yaoqian@sensetime.com>
2021-12-07 10:57:55 +08:00

94 lines
1.9 KiB
C++

// Copyright (c) OpenMMLab. All rights reserved.
#ifndef MMDEPLOY_SRC_EXPERIMENTAL_PIPELINE_IR_H_
#define MMDEPLOY_SRC_EXPERIMENTAL_PIPELINE_IR_H_
#include "core/model.h"
#include "core/module.h"
#include "core/registry.h"
#include "core/status_code.h"
#include "mpl/span.h"
#include "utils/formatter.h"
#if MMDEPLOY_USE_TASKFLOW
#include "taskflow/taskflow.hpp"
#endif
namespace mmdeploy::graph {
using std::pair;
using std::string;
using std::unique_ptr;
using std::vector;
class TaskGraph;
class Node;
class Context {
public:
explicit Context(TaskGraph* graph) : graph_(graph) {}
Value& current() { return context_.back(); }
void push(Value&& ctx) { context_.push_back(std::move(ctx)); }
Value pop() {
auto ctx = std::move(context_.back());
context_.pop_back();
return ctx;
}
size_t size() const noexcept { return context_.size(); }
bool empty() const noexcept { return context_.empty(); }
std::vector<Result<Value>> Execute(Span<std::function<Result<Value>()>> tasks);
private:
vector<Value> context_;
TaskGraph* graph_;
};
class TaskGraph {
friend class Context;
public:
using TaskFunction = std::function<Result<void>(Context& ctx)>;
class Handle {
public:
const std::string& name() const noexcept { return name_; }
void set_name(const std::string& name) { name_ = name; }
private:
std::string name_;
};
~TaskGraph();
Handle* Add(TaskFunction fn);
Result<Value> Run(Value inputs);
private:
std::vector<Result<Value>> Execute(Span<std::function<Result<Value>()>> tasks);
vector<TaskFunction> function_;
vector<unique_ptr<Handle>> handle_;
#if MMDEPLOY_USE_TASKFLOW
tf::Executor executor_;
#endif
// profiling utils
std::vector<double> time_;
int64_t count_{};
};
class Node {
public:
virtual ~Node() = default;
virtual void Build(TaskGraph& graph) = 0;
};
} // namespace mmdeploy::graph
#endif // MMDEPLOY_SRC_EXPERIMENTAL_PIPELINE_IR_H_