| 
									
										
										
										
											2019-04-17 10:49:11 +08:00
										 |  |  | //
 | 
					
						
							|  |  |  | //  ShapeArgMax.cpp
 | 
					
						
							|  |  |  | //  MNN
 | 
					
						
							|  |  |  | //
 | 
					
						
							|  |  |  | //  Created by MNN on 2019/01/10.
 | 
					
						
							|  |  |  | //  Copyright © 2018, Alibaba Group Holding Limited
 | 
					
						
							|  |  |  | //
 | 
					
						
							|  |  |  | 
 | 
					
						
							| 
									
										
										
										
											2020-11-05 16:41:56 +08:00
										 |  |  | #include "shape/SizeComputer.hpp"
 | 
					
						
							| 
									
										
										
										
											2019-12-27 22:16:57 +08:00
										 |  |  | #include "core/Macro.h"
 | 
					
						
							| 
									
										
										
										
											2020-04-29 10:12:16 +08:00
										 |  |  | #include <vector>
 | 
					
						
							| 
									
										
										
										
											2019-04-17 10:49:11 +08:00
										 |  |  | 
 | 
					
						
							|  |  |  | namespace MNN { | 
					
						
							|  |  |  | 
 | 
					
						
							|  |  |  | class ArgMaxComputer : public SizeComputer { | 
					
						
							|  |  |  |     virtual bool onComputeSize(const MNN::Op *op, const std::vector<Tensor *> &inputs, | 
					
						
							|  |  |  |                                const std::vector<Tensor *> &outputs) const override { | 
					
						
							|  |  |  |         MNN_ASSERT(1 == inputs.size()); | 
					
						
							|  |  |  |         MNN_ASSERT(1 == outputs.size()); | 
					
						
							|  |  |  | 
 | 
					
						
							|  |  |  |         // copy dims
 | 
					
						
							|  |  |  |         auto &input       = inputs[0]->buffer(); | 
					
						
							|  |  |  |         auto &output      = outputs[0]->buffer(); | 
					
						
							|  |  |  |         output.dimensions = input.dimensions; | 
					
						
							|  |  |  |         memcpy(output.dim, input.dim, sizeof(halide_dimension_t) * input.dimensions); | 
					
						
							|  |  |  | 
 | 
					
						
							| 
									
										
										
											
												- 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
										 |  |  |         auto argMax = op->main_as_ArgMax(); | 
					
						
							|  |  |  | 
 | 
					
						
							|  |  |  |         const auto inputDimensionFormat = TensorUtils::getDescribe(inputs[0])->dimensionFormat; | 
					
						
							|  |  |  | 
 | 
					
						
							|  |  |  |         TensorUtils::getDescribe(outputs[0])->dimensionFormat = inputDimensionFormat; | 
					
						
							|  |  |  | 
 | 
					
						
							| 
									
										
										
										
											2019-12-27 22:16:57 +08:00
										 |  |  |         if (inputDimensionFormat != MNN_DATA_FORMAT_NC4HW4) { | 
					
						
							| 
									
										
										
											
												- 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
										 |  |  |             int axis = argMax->axis(); | 
					
						
							|  |  |  |             if(axis < 0){ | 
					
						
							|  |  |  |                 axis = input.dimensions + axis; | 
					
						
							|  |  |  |             } | 
					
						
							|  |  |  |             // reduce axis dimension
 | 
					
						
							|  |  |  |             output.dimensions = input.dimensions - 1; | 
					
						
							|  |  |  |             for (int i = 0, j = 0; i < input.dimensions; ++i) { | 
					
						
							|  |  |  |                 if (i == axis) { | 
					
						
							|  |  |  |                     continue; | 
					
						
							|  |  |  |                 } | 
					
						
							|  |  |  |                 output.dim[j].extent = input.dim[i].extent; | 
					
						
							|  |  |  |                 j++; | 
					
						
							|  |  |  |             } | 
					
						
							|  |  |  |             output.dim[input.dimensions - 1].extent = 0; | 
					
						
							|  |  |  |             // set output data type to be INT(according to tensorflow implementation)
 | 
					
						
							|  |  |  |             output.type = halide_type_of<int>(); | 
					
						
							|  |  |  |         } else { | 
					
						
							| 
									
										
										
										
											2020-04-29 10:12:16 +08:00
										 |  |  |             if (argMax->axis() == 0) { | 
					
						
							|  |  |  |                 // Legacy code
 | 
					
						
							|  |  |  |                 // key extent
 | 
					
						
							|  |  |  |                 // really legacy
 | 
					
						
							|  |  |  |                 output.type = halide_type_of<float>(); | 
					
						
							|  |  |  |                 int keyExtent = argMax->topK(); | 
					
						
							|  |  |  |                 if (argMax->outMaxVal()) { | 
					
						
							|  |  |  |                     keyExtent *= 2; | 
					
						
							|  |  |  |                 } | 
					
						
							|  |  |  | 
 | 
					
						
							|  |  |  |                 if (input.dim[3].extent > 1) { | 
					
						
							|  |  |  |                     output.dim[3].extent = keyExtent; | 
					
						
							|  |  |  |                 } else if (input.dim[2].extent > 1) { // iw = ow = 1
 | 
					
						
							|  |  |  |                     output.dim[2].extent = keyExtent; | 
					
						
							|  |  |  |                 } else { // iw = ow = 1, ih = oh = 1;
 | 
					
						
							|  |  |  |                     output.dim[1].extent = keyExtent; | 
					
						
							|  |  |  |                 } | 
					
						
							|  |  |  |             } else { | 
					
						
							|  |  |  |                 TensorUtils::getDescribe(outputs[0])->dimensionFormat = MNN_DATA_FORMAT_NCHW; | 
					
						
							|  |  |  |                 output.type = halide_type_of<float>(); | 
					
						
							|  |  |  |                 int topK = argMax->topK(); | 
					
						
							|  |  |  |                 int axis = argMax->axis(); | 
					
						
							|  |  |  |                 // in caffe, axis may not exist, we set it to 10000 to indicate this situation
 | 
					
						
							|  |  |  |                 // see file: tools/converter/source/caffe/ArgMax.cpp
 | 
					
						
							|  |  |  |                 if (axis != 10000) { | 
					
						
							|  |  |  |                     output.dim[axis].extent = topK; | 
					
						
							|  |  |  |                 } else { | 
					
						
							|  |  |  |                     std::vector<int> outputShape(input.dimensions, 1); | 
					
						
							|  |  |  | 
 | 
					
						
							|  |  |  |                     outputShape[0] = input.dim[0].extent; | 
					
						
							|  |  |  |                     outputShape[2] = topK; | 
					
						
							|  |  |  |                     if (argMax->outMaxVal()) { | 
					
						
							|  |  |  |                         outputShape[1] = 2; | 
					
						
							|  |  |  |                     } | 
					
						
							| 
									
										
										
											
												- 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
										 |  |  | 
 | 
					
						
							| 
									
										
										
										
											2020-04-29 10:12:16 +08:00
										 |  |  |                     for (int ii = 0; ii < outputShape.size(); ii++) { | 
					
						
							|  |  |  |                         output.dim[ii].extent = outputShape[ii]; | 
					
						
							|  |  |  |                     } | 
					
						
							|  |  |  |                 } | 
					
						
							| 
									
										
										
											
												- 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
										 |  |  |             } | 
					
						
							| 
									
										
										
										
											2019-04-17 10:49:11 +08:00
										 |  |  |         } | 
					
						
							|  |  |  | 
 | 
					
						
							|  |  |  |         return true; | 
					
						
							|  |  |  |     } | 
					
						
							|  |  |  | }; | 
					
						
							|  |  |  | 
 | 
					
						
							|  |  |  | REGISTER_SHAPE(ArgMaxComputer, OpType_ArgMax); | 
					
						
							| 
									
										
										
										
											2019-12-27 22:16:57 +08:00
										 |  |  | REGISTER_SHAPE(ArgMaxComputer, OpType_ArgMin); | 
					
						
							| 
									
										
										
										
											2019-04-17 10:49:11 +08:00
										 |  |  | 
 | 
					
						
							|  |  |  | } // namespace MNN
 |