| 
									
										
										
										
											2019-04-17 10:49:11 +08:00
										 |  |  | //
 | 
					
						
							|  |  |  | //  ShapeTranspose.cpp
 | 
					
						
							|  |  |  | //  MNN
 | 
					
						
							|  |  |  | //
 | 
					
						
							|  |  |  | //  Created by MNN on 2019/01/10.
 | 
					
						
							|  |  |  | //  Copyright © 2018, Alibaba Group Holding Limited
 | 
					
						
							|  |  |  | //
 | 
					
						
							|  |  |  | 
 | 
					
						
							| 
									
										
										
										
											2019-12-27 22:16:57 +08:00
										 |  |  | #include "core/Macro.h"
 | 
					
						
							|  |  |  | #include "core/SizeComputer.hpp"
 | 
					
						
							|  |  |  | #include "core/TensorUtils.hpp"
 | 
					
						
							| 
									
										
										
										
											2019-04-17 10:49:11 +08:00
										 |  |  | namespace MNN { | 
					
						
							|  |  |  | 
 | 
					
						
							|  |  |  | class TransposeComputer : public SizeComputer { | 
					
						
							|  |  |  |     virtual bool onComputeSize(const MNN::Op* op, const std::vector<Tensor*>& inputs, | 
					
						
							|  |  |  |                                const std::vector<Tensor*>& outputs) const override { | 
					
						
							|  |  |  |         const Tensor* input = inputs[0]; | 
					
						
							|  |  |  |         Tensor* perm        = inputs[1]; | 
					
						
							|  |  |  |         const int dims = input->buffer().dimensions; | 
					
						
							|  |  |  |         MNN_ASSERT(dims == perm->buffer().dim[0].extent); | 
					
						
							|  |  |  | 
 | 
					
						
							|  |  |  |         std::vector<int32_t> permutation; | 
					
						
							| 
									
										
										
										
											2019-08-22 20:13:46 +08:00
										 |  |  |         if (perm->getType().code == halide_type_int && 32 == perm->getType().bits) { | 
					
						
							| 
									
										
										
										
											2019-04-17 10:49:11 +08:00
										 |  |  |             for (int i = 0; i < perm->buffer().dim[0].extent; i++) { | 
					
						
							|  |  |  |                 permutation.push_back(perm->host<int32_t>()[i]); | 
					
						
							|  |  |  |             } | 
					
						
							|  |  |  |         } else { | 
					
						
							|  |  |  |             MNN_ASSERT(false); | 
					
						
							|  |  |  |         } | 
					
						
							|  |  |  | 
 | 
					
						
							|  |  |  |         outputs[0]->buffer().dimensions = dims; | 
					
						
							| 
									
										
											  
											
												- 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
										 |  |  |         outputs[0]->buffer().type = input->getType(); | 
					
						
							| 
									
										
										
										
											2019-04-17 10:49:11 +08:00
										 |  |  |         for (int i = 0; i < dims; ++i) { | 
					
						
							|  |  |  |             const int32_t d                    = permutation[i]; | 
					
						
							|  |  |  |             outputs[0]->buffer().dim[i].extent = input->buffer().dim[d].extent; | 
					
						
							|  |  |  |         } | 
					
						
							| 
									
										
										
										
											2019-07-02 18:01:08 +08:00
										 |  |  |         TensorUtils::getDescribe(outputs[0])->dimensionFormat = TensorUtils::getDescribe(inputs[0])->dimensionFormat; | 
					
						
							| 
									
										
										
										
											2019-04-17 10:49:11 +08:00
										 |  |  | 
 | 
					
						
							|  |  |  |         return true; | 
					
						
							|  |  |  |     } | 
					
						
							|  |  |  | }; | 
					
						
							|  |  |  | 
 | 
					
						
							| 
									
										
										
										
											2019-08-22 20:13:46 +08:00
										 |  |  | REGISTER_SHAPE_INPUTS(TransposeComputer, OpType_Transpose, {1}); | 
					
						
							| 
									
										
										
										
											2019-04-17 10:49:11 +08:00
										 |  |  | } // namespace MNN
 |