MNN/source/backend/opencl/execution/buffer/DepthwiseConvBufExecution.hpp

45 lines
1.5 KiB
C++

//
// DepthwiseConvBufExecution.hpp
// MNN
//
// Created by MNN on 2019/01/31.
// Copyright © 2018, Alibaba Group Holding Limited
//
#ifndef MNN_OPENCL_BUFFER_CLOSED
#ifndef DepthwiseConvBufExecution_hpp
#define DepthwiseConvBufExecution_hpp
#include "ConvBufExecution.hpp"
namespace MNN {
namespace OpenCL {
class DepthwiseConvBufExecution : public ConvBufCommonExecution {
public:
DepthwiseConvBufExecution(const std::vector<Tensor *> &inputs, const MNN::Op *op, Backend *backend);
virtual ~DepthwiseConvBufExecution();
virtual ErrorCode onResize(const std::vector<Tensor *> &inputs, const std::vector<Tensor *> &outputs) override;
virtual ErrorCode onExecute(const std::vector<Tensor *> &inputs, const std::vector<Tensor *> &outputs) override;
std::pair<std::vector<uint32_t>, int> DepthwiseConvBufLwsTune(const cl::Kernel &kernel, const std::vector<uint32_t> &gws, const std::string &kernelName, const uint32_t maxWorkGroupSize);
private:
const Convolution2DCommon *mConv2dCommonParams;
const Convolution2D *mCon2dParams;
std::vector<int> mStrides{1, 1};
std::vector<int> mPaddings{0, 0};
std::vector<int> mDilations{1, 1};
std::shared_ptr<Tensor> mFilter;
cl::Kernel mKernel;
uint32_t mMaxWorkGroupSize;
std::vector<uint32_t> mGlobalWorkSize{1, 1, 1};
std::vector<uint32_t> mLocalWorkSize{1, 1, 1, 1};
bool mStride_1 = false;
std::set<std::string> mBuildOptions;
};
} // namespace OpenCL
} // namespace MNN
#endif /* DepthwiseConvBufExecution_hpp */
#endif /* MNN_OPENCL_BUFFER_CLOSED */