- dynamic computation graph (beta)
- add supports (/express)
- add tests
- add benchmarks with it (/benchmark/exprModels)
- Python
- MNN engine and tools were submitted to pip
- available on Windows/macOS/Linux
- Engine/Converter
- add supports for each op benchmarking
- refactor optimizer by separating steps
- CPU
- add supports for Conv3D, Pool3D, ELU, ReverseSequence
- fix ArgMax, Permute, Scale, BinaryOp, Slice, SliceTf
- OpenCL
- add half transform in CPU
- add broadcast supports for binary
- optimize Conv2D, Reshape, Eltwise, Gemm, etc.
- OpenGL
- add sub, real div supports for binary
- add supports for unary
- optimize Conv2D, Reshape
- Vulkan
- add max supports for eltwise
- Metal
- fix metallib missing problem
- Train/Quantization
- use express to refactor training codes
2019-09-26 21:02:07 +08:00
|
|
|
//
|
|
|
|
// Utils.hpp
|
|
|
|
// MNN
|
|
|
|
//
|
|
|
|
// Created by MNN on 2019/07/26.
|
|
|
|
// Copyright © 2018, Alibaba Group Holding Limited
|
|
|
|
//
|
|
|
|
|
2020-01-15 13:33:47 +08:00
|
|
|
#ifndef Utils_hpp
|
|
|
|
#define Utils_hpp
|
2019-12-27 22:16:57 +08:00
|
|
|
#include <MNN/expr/Expr.hpp>
|
2020-02-26 23:08:52 +08:00
|
|
|
#include "Type_generated.h"
|
2020-12-14 18:11:56 +08:00
|
|
|
#include "MNN_generated.h"
|
2020-01-15 13:33:47 +08:00
|
|
|
#include <MNN/expr/Executor.hpp>
|
2021-11-30 10:10:53 +08:00
|
|
|
#include "core/AutoStorage.h"
|
- dynamic computation graph (beta)
- add supports (/express)
- add tests
- add benchmarks with it (/benchmark/exprModels)
- Python
- MNN engine and tools were submitted to pip
- available on Windows/macOS/Linux
- Engine/Converter
- add supports for each op benchmarking
- refactor optimizer by separating steps
- CPU
- add supports for Conv3D, Pool3D, ELU, ReverseSequence
- fix ArgMax, Permute, Scale, BinaryOp, Slice, SliceTf
- OpenCL
- add half transform in CPU
- add broadcast supports for binary
- optimize Conv2D, Reshape, Eltwise, Gemm, etc.
- OpenGL
- add sub, real div supports for binary
- add supports for unary
- optimize Conv2D, Reshape
- Vulkan
- add max supports for eltwise
- Metal
- fix metallib missing problem
- Train/Quantization
- use express to refactor training codes
2019-09-26 21:02:07 +08:00
|
|
|
namespace MNN {
|
2022-12-30 15:18:58 +08:00
|
|
|
class Session;
|
- dynamic computation graph (beta)
- add supports (/express)
- add tests
- add benchmarks with it (/benchmark/exprModels)
- Python
- MNN engine and tools were submitted to pip
- available on Windows/macOS/Linux
- Engine/Converter
- add supports for each op benchmarking
- refactor optimizer by separating steps
- CPU
- add supports for Conv3D, Pool3D, ELU, ReverseSequence
- fix ArgMax, Permute, Scale, BinaryOp, Slice, SliceTf
- OpenCL
- add half transform in CPU
- add broadcast supports for binary
- optimize Conv2D, Reshape, Eltwise, Gemm, etc.
- OpenGL
- add sub, real div supports for binary
- add supports for unary
- optimize Conv2D, Reshape
- Vulkan
- add max supports for eltwise
- Metal
- fix metallib missing problem
- Train/Quantization
- use express to refactor training codes
2019-09-26 21:02:07 +08:00
|
|
|
namespace Express {
|
2020-01-15 13:33:47 +08:00
|
|
|
struct Expr::Inside {
|
2020-11-05 16:41:56 +08:00
|
|
|
Inside(int outputSize);
|
2021-04-08 15:34:23 +08:00
|
|
|
Inside(Tensor* tensor, bool own = false);
|
2020-11-05 16:41:56 +08:00
|
|
|
~ Inside();
|
2020-01-15 13:33:47 +08:00
|
|
|
std::vector<Variable::Info> mOutputInfos;
|
2020-11-05 16:41:56 +08:00
|
|
|
std::vector<Tensor*> mOutputTensors;
|
2020-01-15 13:33:47 +08:00
|
|
|
Executor::Requirement mReq;
|
|
|
|
std::shared_ptr<Executor::ComputeCache> mCache;
|
2020-02-26 23:08:52 +08:00
|
|
|
int mCacheOffset = 0;
|
2020-02-28 17:26:43 +08:00
|
|
|
bool mInfoDirty = true;
|
2020-03-03 06:55:38 +08:00
|
|
|
bool mContentDirty = true;
|
2021-01-06 16:29:37 +08:00
|
|
|
bool mOwnTensor = true;
|
|
|
|
Tensor* mHostTensor = nullptr;
|
2021-11-30 10:10:53 +08:00
|
|
|
std::shared_ptr<Backend> mHoldBackend;
|
|
|
|
};
|
|
|
|
struct Executor::DebugTools {
|
|
|
|
TensorCallBackWithInfo before = nullptr;
|
|
|
|
TensorCallBackWithInfo after = nullptr;
|
2022-07-22 09:59:30 +08:00
|
|
|
mutable float flops = 0.0f;
|
2020-01-15 13:33:47 +08:00
|
|
|
};
|
2022-12-30 15:18:58 +08:00
|
|
|
struct Executor::SubGraph {
|
|
|
|
std::unique_ptr<MNN::SubGraphProtoT> info;
|
|
|
|
std::vector<std::string> depends;
|
|
|
|
};
|
|
|
|
class Executor::ComputeCache {
|
|
|
|
public:
|
|
|
|
void setContentDirty();
|
|
|
|
void* mapOutput(int offset, Tensor* dest);
|
|
|
|
|
|
|
|
~ ComputeCache();
|
|
|
|
ComputeCache() {
|
|
|
|
// Do nothing
|
|
|
|
}
|
|
|
|
|
|
|
|
ErrorCode compute();
|
|
|
|
ErrorCode resize();
|
|
|
|
ErrorCode resizeImpl();
|
|
|
|
Session* getSession() {
|
|
|
|
return mSession.get();
|
|
|
|
}
|
|
|
|
friend class Executor;
|
|
|
|
private:
|
|
|
|
std::set<std::shared_ptr<Expr::Inside>> mInputInside;
|
|
|
|
std::set<std::shared_ptr<ComputeCache>> mInputs;
|
|
|
|
std::shared_ptr<Session> mSession;
|
|
|
|
bool mContentDirty = true;
|
|
|
|
bool mShapeDirty = true;
|
|
|
|
std::vector<std::shared_ptr<BufferStorage>> mCacheBuffers;
|
|
|
|
#ifdef MNN_EXPRESS_MEMLEAK_DEBUG
|
|
|
|
static int gInstanceCount;
|
|
|
|
#endif
|
|
|
|
};
|
- dynamic computation graph (beta)
- add supports (/express)
- add tests
- add benchmarks with it (/benchmark/exprModels)
- Python
- MNN engine and tools were submitted to pip
- available on Windows/macOS/Linux
- Engine/Converter
- add supports for each op benchmarking
- refactor optimizer by separating steps
- CPU
- add supports for Conv3D, Pool3D, ELU, ReverseSequence
- fix ArgMax, Permute, Scale, BinaryOp, Slice, SliceTf
- OpenCL
- add half transform in CPU
- add broadcast supports for binary
- optimize Conv2D, Reshape, Eltwise, Gemm, etc.
- OpenGL
- add sub, real div supports for binary
- add supports for unary
- optimize Conv2D, Reshape
- Vulkan
- add max supports for eltwise
- Metal
- fix metallib missing problem
- Train/Quantization
- use express to refactor training codes
2019-09-26 21:02:07 +08:00
|
|
|
class Utils {
|
|
|
|
public:
|
|
|
|
static void copyInfoToTensor(Tensor* dest, const Variable::Info* source);
|
|
|
|
static void copyTensorToInfo(Variable::Info* dest, const Tensor* source);
|
2020-02-26 23:08:52 +08:00
|
|
|
static DataType convertDataType(halide_type_t type);
|
- build:
- unify schema building in core and converter;
- add more build script for android;
- add linux build script for python;
- ops impl:
- add floor mod support in binary;
- use eltwise impl in add/max/sub/mul binary for optimization;
- remove fake double support in cast;
- fix 5d support for concat;
- add adjX and adjY support for batch matmul;
- optimize conv2d back prop filter;
- add pad mode support for conv3d;
- fix bug in conv2d & conv depthwise with very small feature map;
- optimize binary without broacast;
- add data types support for gather;
- add gather ND support;
- use uint8 data type in gather v2;
- add transpose support for matmul;
- add matrix band part;
- add dim != 4 support for padding, reshape & tensor convert;
- add pad type support for pool3d;
- make ops based on TensorFlow Lite quantization optional;
- add all & any support for reduction;
- use type in parameter as output type in reduction;
- add int support for unary;
- add variable weight support for conv2d;
- fix conv2d depthwise weights initialization;
- fix type support for transpose;
- fix grad outputs count for reduce grad and reshape grad;
- fix priorbox & detection output;
- fix metal softmax error;
- python:
- add runSessionWithCallBackInfo interface;
- add max nodes limit (1400) for visualization tool;
- fix save error in python3;
- align default dim;
- convert:
- add extra design for optimization;
- add more post converting optimizers;
- add caffe v1 weights blob support;
- add cast, unary, conv transpose support for onnx model;
- optimize batchnorm, conv with variable weights, prelu, reshape, slice, upsample for onnx model;
- add cos/sin/atan/tan support for unary for tensorflow model;
- add any/all support for reduction for tensorflow model;
- add elu, conv3d, pool3d support for tensorflow model;
- optimize argmax, batchnorm, concat, batch to space, conv with variable weights, prelu, slice for tensorflow model;
- others:
- fix size computer lock;
- fix thread pool deadlock;
- add express & parameters in express;
- rewrite blitter chooser without static map;
- add tests for expr;
2019-10-29 13:37:26 +08:00
|
|
|
static int convertFormat(Dimensionformat format);
|
2019-12-27 22:16:57 +08:00
|
|
|
static Express::Dimensionformat revertFormat(int format);
|
2020-02-26 23:08:52 +08:00
|
|
|
static halide_type_t revertDataType(DataType dataType);
|
2020-11-05 16:41:56 +08:00
|
|
|
static bool allocMemoryForHostTensor(Tensor* dest);
|
|
|
|
static bool releaseMemoryForHostTensor(Tensor* dest);
|
2022-07-22 09:59:30 +08:00
|
|
|
static Tensor* getTensor(VARP var);
|
2022-12-30 15:18:58 +08:00
|
|
|
static EXPRP makeRaster(const std::vector<VARP>& vars, const std::vector<int>& regions, const std::vector<int>& shape, halide_type_t dataType, MNN_DATA_FORMAT format);
|
- dynamic computation graph (beta)
- add supports (/express)
- add tests
- add benchmarks with it (/benchmark/exprModels)
- Python
- MNN engine and tools were submitted to pip
- available on Windows/macOS/Linux
- Engine/Converter
- add supports for each op benchmarking
- refactor optimizer by separating steps
- CPU
- add supports for Conv3D, Pool3D, ELU, ReverseSequence
- fix ArgMax, Permute, Scale, BinaryOp, Slice, SliceTf
- OpenCL
- add half transform in CPU
- add broadcast supports for binary
- optimize Conv2D, Reshape, Eltwise, Gemm, etc.
- OpenGL
- add sub, real div supports for binary
- add supports for unary
- optimize Conv2D, Reshape
- Vulkan
- add max supports for eltwise
- Metal
- fix metallib missing problem
- Train/Quantization
- use express to refactor training codes
2019-09-26 21:02:07 +08:00
|
|
|
};
|
2025-07-23 14:10:58 +08:00
|
|
|
class Executor::RuntimeExecuteWrap {
|
|
|
|
public:
|
|
|
|
RuntimeExecuteWrap(const RuntimeInfo& info);
|
|
|
|
~ RuntimeExecuteWrap();
|
|
|
|
private:
|
|
|
|
const RuntimeInfo& mRt;
|
|
|
|
};
|
- dynamic computation graph (beta)
- add supports (/express)
- add tests
- add benchmarks with it (/benchmark/exprModels)
- Python
- MNN engine and tools were submitted to pip
- available on Windows/macOS/Linux
- Engine/Converter
- add supports for each op benchmarking
- refactor optimizer by separating steps
- CPU
- add supports for Conv3D, Pool3D, ELU, ReverseSequence
- fix ArgMax, Permute, Scale, BinaryOp, Slice, SliceTf
- OpenCL
- add half transform in CPU
- add broadcast supports for binary
- optimize Conv2D, Reshape, Eltwise, Gemm, etc.
- OpenGL
- add sub, real div supports for binary
- add supports for unary
- optimize Conv2D, Reshape
- Vulkan
- add max supports for eltwise
- Metal
- fix metallib missing problem
- Train/Quantization
- use express to refactor training codes
2019-09-26 21:02:07 +08:00
|
|
|
} // namespace Express
|
|
|
|
} // namespace MNN
|
2020-01-15 13:33:47 +08:00
|
|
|
#endif
|