Paddle

Commit Graph

Author	SHA1	Message	Date
fengjiayi	ae6009dfef	fix compile error	8 years ago
fengjiayi	77af58f8f7	Change gradient Op registry mechanism OLD: op_type -> grad_op_creator NEW: grad_op_type -> grad_op_creator op_type -> grad_op_type	8 years ago
fengjiayi	f4e25550cd	Fix compile error Replace `OperatorPtr` with `std::shared_ptr<OperatorBase>`	8 years ago
fengjiayi	5f3bc2a44a	Merge branch 'develop' of https://github.com/PaddlePaddle/Paddle into backward	8 years ago
Yu Yang	c2543f5b29	Remove ScopePtr and OperatorPtr * ScopePtr means pointer of scope, but it can be shared or uniqued. Change it to std::shared_ptr<Scope> to make code better to read.	8 years ago
fengjiayi	9418717f80	Fix compile errors	8 years ago
fengjiayi	99a5904c75	Merge branch 'backward_dev' into backward	8 years ago
fengjiayi	8a5ee462a5	Fix some compile errors	8 years ago
dongzhihong	81a352aff7	"test fc without gradient"	8 years ago
dongzhihong	14424f314c	"use built-in operator"	8 years ago
fengjiayi	e192d0fd01	Refactor the implementation of gradient Op creating	8 years ago
fengjiayi	3dc70ff282	Merge branch 'backward' of https://github.com/dzhwinter/Paddle into backward	8 years ago
dongzhihong	8bc4892106	"fix comment "	8 years ago
fengjiayi	94a6b1fa43	rename a macro	8 years ago
dongzhihong	cb95587feb	"ignore some gradient of specific op"	8 years ago
fengjiayi	bf4da3d97a	Refactor Rigistry::CreateGradOp() We put forward Op's inputs, outputs and output gradients into Grad Op's inputs, and put forward Op's input gradients into Grad Op's output. So Grad Op's `in_out_idx`, `input_format` and 'output format' need to be rebuilt during Op creating.	8 years ago
fengjiayi	7f1533f251	test collaborating	8 years ago
dongzhihong	9890b23c82	fix confilct	8 years ago
dongzhihong	816b4c8ab0	"add backward Op"	8 years ago
Yu Yang	f6a51d9b4e	Make CreateOp in Plain C++ params	8 years ago
Yu Yang	c1219a530c	Change `in_out_idxs_` to shared_ptr * `in_out_idxs_` shares between all operator instance in same type of operator.	8 years ago
Qiao Longfei	051676a7e4	support multiple template parameter in KernelType for REGISTER_OP_XPU_KERNEL (#2932 )	8 years ago
Yu Yang	0e77b31a02	Merge branch 'develop' into feature/op_creation_methods	8 years ago
Yan Chunwei	a0caf23430	Op varient inputs (#2901 ) * add inputs * add ut for multiple inputs * fix AddToLayer * op_desc -> op_proto * CreateArgumentOffsetMap -> CreateInOutOffsetMap * move CreateInOutOffsetMap from OperatorBase to op registry * arg_idxs_ -> in_out_idxs_	8 years ago
Qiao Longfei	80a26a6308	check duplicate of ProtoAndCheckerMaker (#2903 )	8 years ago
Yu Yang	c5bc126762	Follow comment, rename to `GenerateTempVariableName`	8 years ago
Yu Yang	68f6ac011c	Merge branch 'develop' into feature/op_creation_methods	8 years ago
Qiao Longfei	45ce1649a1	change net to operator (#2846 ) * OperatorBase should not store OpDesc because not All op contains an OpDesc and not all ops create from OpDesc. * Networks do not contain OpDesc and are not created by OpDesc * Do not register Network to OpRegistry. * The network is directly created by the user in Python. Not from registry. * Correctly handle the `inputs` and `outputs` of a Network. * Add CompleteAddOp() methods * Remove `AddOp(OpDesc&)` in net-op. All op are added by OperatorPtr. * Rewrite unit test for truly tested what networks do. * optimise operator_test	8 years ago
Qiao Longfei	afa99d9ae6	add ADD_OP_CPU to enable add op with only cpu kernel (#2896 ) * add ADD_OP_CPU to enable add op with only cpu kernel	8 years ago
Yu Yang	9e0c6800c5	Python Generate OpCreation Methods by OpProto All OpCreation method are generated by `create_op_creation_methods::__bootstrap__` method, and stores in `op_creations` object and its methods. There are three parts to implement this feature. 1. Get all registered `OpProto` from C++ side. It is implemented in `get_all_op_protos` method. 1. Create a function to convert `kwargs` to `OpDesc` base on each op's `OpProto`. The `OpDescCreationMethod` class. 1. Convert `OpProto` to `docstring` by `get_docstring_from_op_proto` method. All three methods are unit tested. The `__bootstrap__` just combines them together and create a method in runtime. For details, please reference the doc string in `create_op_creation_methods.py` and the unit test `test_op_creation_methods.py`.	8 years ago
Yu Yang	1faf5e0313	Merge pull request #2864 from reyoung/feature/op_creation_methods Get OpProtos in Python	8 years ago
Qiao Longfei	58f3de95cf	Optimize ptr (#2851 ) * use OperatorPtr = std::shared_ptr<OperatorBase>; * use ScopePtr = std::share_ptr<Scope>;	8 years ago
Yu Yang	2462d0c5fe	Let OpProto support multiple and temporary (#2860 ) * Let OpProto support multiple and temporary * Each input/output of Paddle's Op could be a list. Add multiple mark to OpProto. Also add a `input_format`/`output_format` attribute if that Op has multiple input or output. The format of that attribute please reference the comments in `op_proto.proto` * Add temporary mark, because some output of an Op is not used by user but used by other op for faster computation. Explicitly mark which output is temporary could let future memory/computation optimization. * Add generated field to AttrProto. * Add `AddInputs`/`AddOutputs` function * It is more readable to invoke `AddInputs` not `AddInput(multiple=true)`.	8 years ago
Yu Yang	cd5113c197	Init commit	8 years ago
Yu Yang	79b70c2d69	Follow comments * Convert `op` --> `operators` * Remove AddType in OpProtoMaker, because type is part of registry. * Rename CPU_OR_GPU --> DEVICE_TYPE in registry macro.	8 years ago
Yu Yang	a0aaafe9de	Add a sample op, `add_op` * Refine register methods, make Op can get rid of whole-archieve * `USE_OP` before a op is used. * Add unittest for add_op.	8 years ago
Qiao Longfei	728665d709	Add Init to OperatorBase (#2838 )	8 years ago
Qiao Longfei	be441f7d16	test OpKernel (#2820 ) Add unit test for OpKernel	8 years ago
Qiao Longfei	a2e5f652d3	add operator base (#2725 ) Add OperatorBase. issue: https://github.com/PaddlePaddle/Paddle/issues/2790 Paddle design the Operator with Kernel. OperatorBase has no type and device information when create, One operator can have multiple kernels, Operator will choose a kernel to run according to context. The kernel should be bind to Operator before or during Operator running.	8 years ago
Yu Yang	267f9a2cdf	Move static variable defined in .cc (#2782 ) * Move static variable defined in .cc We cannot define static variable in .h, because it will be multi-defined errors. Also fix some cpp syntax, like: * Prefer to use algorithm not manually for-loop, to make code more readable. * Remove unused `()`. * Enforce take a bool. It is no need `xxx==true`. * Use range-based for-loop iterator from op_desc.attrs * Fix a protential static variable init order error	8 years ago
fengjiayi	1d2ef1db82	[draft] add registry for Op, OpProto and OpAttrChecker (#2739 ) * init op_registry.h * dev op_registry.h * add 'attr_checker.h', which is a draft of op attribute checker. * rename some macro parameters * 1. Use `Attribute` and `AttributeMap` instead of `OpDesc`. `AttributeMap` is a unordered_map of <string, Attribute>, and `Attribute` is a boost::variant object to hold multiple types of attribute value. 2. Use `PADDLE_ENFORCE` to print checkers' fail message. 3. Abstract default value operations to a new function: `DefaultChecker`. * rename DefaultChecker to DefaultValueSetter ZZ * Finish op_registry 1. Complete the development of interfaces between OpRegistry and Protobuf. 2. Add unit test for op_registry.h * Add demo and test of custome checker * fix merge conflict	8 years ago

41 Commits (358261f0bdf2ce887a3ff77218694828a6527ede)