mmdeploy/csrc/core/tensor.h
lzhangzz 640aa03538
Support Windows (#106)
* minor changes

* support windows

* fix GCC build

* fix lint

* reformat

* fix Windows build

* fix GCC build

* search backend ops for onnxruntime

* fix lint

* fix lint

* code clean-up

* code clean-up

* fix clang build

* fix trt support

* fix cmake for ncnn

* fix cmake for openvino

* fix SDK Python API

* handle ops for other backends (ncnn, trt)

* handle SDK Python API library location

* robustify linkage

* fix cuda

* minor fix for openvino & ncnn

* use CMAKE_CUDA_ARCHITECTURES if set

* fix cuda preprocessor

* fix misc

* fix pplnn & pplcv, drop support for pplcv<0.6.0

* robustify cmake

* update build.md (#2)

* build dynamic modules as module library & fix demo (partially)

* fix candidate path for mmdeploy_python

* move "enable CUDA" to cmake config for demo

* refine demo cmake

* add comment

* fix ubuntu build

* revert docs/en/build.md

* fix C API

* fix lint

* Windows build doc (#3)

* check in docs related to mmdeploy build on windows

* update build guide on windows platform

* update build guide on windows platform

* make path of thirdparty libraries consistent

* make path consistency

* correct build command for custom ops

* correct build command for sdk

* update sdk build instructions

* update doc

* correct build command

* fix lint

* correct build command and fix lint

Co-authored-by: lvhan <lvhan@pjlab.org>

* trailing whitespace (#4)

* minor fix

* fix sr sdk model

* fix type deduction

* fix cudaFree after driver shutting down

* update ppl.cv installation warning (#5)

* fix device allocator threshold & fix lint

* update doc (#6)

* update ppl.cv installation warning

* missing 'git clone'

Co-authored-by: chenxin <chenxin2@sensetime.com>
Co-authored-by: zhangli <zhangli@sensetime.com>
Co-authored-by: lvhan028 <lvhan_028@163.com>
Co-authored-by: lvhan <lvhan@pjlab.org>
2022-02-24 20:08:44 +08:00

82 lines
1.8 KiB
C++

// Copyright (c) OpenMMLab. All rights reserved.
#ifndef CORE_TENSOR_H
#define CORE_TENSOR_H
#include <string>
#include <vector>
#include "device.h"
#include "types.h"
namespace mmdeploy {
using TensorShape = std::vector<int64_t>;
struct TensorDesc {
Device device;
DataType data_type{DataType::kFLOAT};
TensorShape shape;
std::string name;
};
class MMDEPLOY_API Tensor {
public:
Tensor() = default;
Tensor(const Tensor&) = default;
Tensor(Tensor&&) noexcept = default;
Tensor& operator=(const Tensor&) = default;
Tensor& operator=(Tensor&&) noexcept = default;
Tensor(const TensorDesc& desc, Allocator allocator = {}); // NOLINT
Tensor(const TensorDesc& desc, Buffer buffer);
Tensor(const TensorDesc& desc, std::shared_ptr<void> data);
~Tensor() = default;
const TensorDesc& desc() const;
const TensorShape& shape() const;
TensorShape::value_type shape(int dim) const;
DataType data_type() const;
const char* name() const;
int64_t size() const;
int64_t byte_size() const;
const Buffer& buffer() const;
Buffer& buffer();
Device device() const;
void Reshape(const TensorShape& shape);
Tensor Slice(int index);
Result<void> CopyFrom(const Tensor& tensor, Stream stream = {});
Result<void> CopyTo(Tensor& tensor, Stream stream = {}) const;
Result<void> CopyFrom(void* host_ptr, Stream stream = {});
Result<void> CopyTo(void* host_ptr, Stream stream = {}) const;
Allocator allocator() { return allocator_; }
template <typename T = void>
T* data() {
return GetNative<T*>(buffer());
}
template <typename T = void, typename U = std::add_const_t<T> >
U* data() const {
return GetNative<U*>(buffer());
}
private:
void Allocate();
TensorDesc desc_;
Allocator allocator_;
Buffer buffer_;
};
// static_assert(sizeof(Tensor) == 80);
} // namespace mmdeploy
#endif // !CORE_TENSOR_H