203 Commits (4a83dde594d0aa6d19aeff7471b040277a8a839f)

Author	SHA1	Message	Date
Yu Yang	f0a85b0805	Rename __is_close -> assert_is_close()	8 years ago
Yu Yang	c957445c72	A better error message for gradient checker ... * Give which parameter, which element are wrong. And what max_diff is.	8 years ago
dongzhihong	bbd7378b43	"ci job failed weired. restart ci job."	8 years ago
dongzhihong	6bac3e17b5	"remove unused test net modified"	8 years ago
dongzhihong	d98e299d3b	"keep same with uniform random op"	8 years ago
dongzhihong	b7ee1e7d9c	"backward check todo"	8 years ago
dongzhihong	6535a7b001	Merge remote-tracking branch 'origin/develop' into random_op	8 years ago
Qiao Longfei	e31a469ee0	add gradient test framework (#3226) ... * init grad op checker * can run * add GradeChecker class * use get_numeric_gradient * refine code * add softmax and cross entropy auto grad test * use close to judge op_grad and numeric_grad * add cpu and gpu compare * add comments * add support_gpu * fix allclose * fix name error and symplify code * optimize gradient checker * add test_cross_entropy_op * update gradient_checker.py * optimize code * use random.uniform instead of random.random * fix type bug * optimize check_grad * put SupportGPU into OperatorBase * typo	8 years ago
dongzhihong	23ac845921	Merge remote-tracking branch 'origin/develop' into random_op	8 years ago
superjom	e9a92e3ed8	add fc test	8 years ago
superjom	001b62a4c7	finish simple rnn in python	8 years ago
dongzhihong	8804b242d8	Merge remote-tracking branch 'origin/develop' into random_op	8 years ago
dongzhihong	52d2ebdaef	"test gaussian random in python side"	8 years ago
Yu Yang	b861c01963	Merge branch 'develop' into feature/uniform_random_op	8 years ago
fengjiayi	e7b3e13959	Rewrite test based on new python test format	8 years ago
fengjiayi	5d2b3fb09d	Merge branch 'develop' of https://github.com/PaddlePaddle/Paddle into dev_add_FillZerosLikeOp_test	8 years ago
Yu Yang	ffd6c3cea6	Merge branch 'develop' into feature/change_op_creation	8 years ago
Yu Yang	e5847924f3	Merge branch 'develop' of github.com:baidu/Paddle into feature/uniform_random_op	8 years ago
Qiao Longfei	493396d81c	add support_gpu (#3304) ... * add support_gpu * fix allclose * fix name error and symplify code	8 years ago
dongzhihong	58561d8f56	Merge remote-tracking branch 'origin/develop' into random_op	8 years ago
Yu Yang	d7f0eb6baf	Merge branch 'feature/change_op_creation' into feature/uniform_random_op	8 years ago
Yu Yang	460326f4b2	Fix CI Test	8 years ago
Yi Wang	efa50a4525	Merge pull request #3181 from qingqing01/sigmoid_bp ... Sigmoid backward implementation.	8 years ago
Yu Yang	9f816352e5	Follow comments	8 years ago
Yan Chunwei	7665bdba13	Rnn forward logic test (#3291) ... * finish forward debug	8 years ago
Yu Yang	5d074c917c	Merge branch 'develop' into feature/change_op_creation	8 years ago
gangliao	ec2c753c3d	Merge pull request #3290 from PaddlePaddle/cblas_new ... Import HPC linear algebra libs as cblas target for cc_library deps	8 years ago
Yu Yang	fd0bdb4f8a	Merge branch 'feature/change_op_creation' into feature/uniform_random_op	8 years ago
dangqingqing	9f490c7748	update to develop branch.	8 years ago
liaogang	7ecdf6ad93	Add py_test	8 years ago
Yu Yang	e376bda42c	Add uniform random operator ... It can be run both CPU/GPU. configure attributes are: * min: the min value of uniform random * max: the max value of uniform random * dims: the dimension of output tensor * seed: the random seed of uniform random. 0 means generate a seed each time.	8 years ago
dangqingqing	f72cefabce	fix typo and resolve conflicts.	8 years ago
Yu Yang	a1e16bb5d3	Merge branch 'develop' into feature/change_op_creation	8 years ago
dongzhihong	2447c34a0e	merge origin/develop	8 years ago
Yan Chunwei	c72253127f	crossentropy grad op (#3186) ... * init cross entropy graident * add crossentropy grad op * remove details * fix static compile	8 years ago
QI JUN	10ee3dcbcf	Merge pull request #3248 from QiJune/refine_op_test_compare ... use isclose instead of assert_almost_equal	8 years ago
Yu Yang	b4ebb3c85c	Change attr_type_pb2 to attribute_pb2 ... Make ci pass	8 years ago
fengjiayi	9b0fce516a	Merge branch 'develop' of https://github.com/PaddlePaddle/Paddle into dev_add_FillZerosLikeOp_test	8 years ago
qijun	dcdd2903fe	use isclose instead of assert_almost_equal	8 years ago
dangqingqing	7fdfe1a9c7	Merge branch 'develop' of https://github.com/PaddlePaddle/Paddle into unit_test	8 years ago
dangqingqing	c540aa040f	Refine unit test in op_test_util	8 years ago
Yu Yang	89d33ff83f	Complete chagne op creation method. ... Currently use `Operator("fc", X="x", W='w1', B='b1')` as operator creation method. Fix #3198	8 years ago
Yu Yang	a239418bdf	Fix unittest for operator.py ... Rename operator.py to op.py because it is conflict with protobuf	8 years ago
Yu Yang	53f85df1ab	Start doing `python.framework.operator`	8 years ago
Yu Yang	aaf2d66db8	Merge branch 'develop' of github.com:baidu/Paddle into gradient_check_utils	8 years ago
fengjiayi	df84e39a00	Merge branch 'develop' of https://github.com/PaddlePaddle/Paddle into dev_add_FillZerosLikeOp_test	8 years ago
Qiao Longfei	d953611e86	Softmax grad op (#3164) ... * init softmax grad op * add compute code * export Backward to python * update test ,export op.type to python * update python test, fix compute bug * update unit test * use eigen * optimize eigen code * add gpu test * register softmax_grad GPU kernel and fix test bug * typo * follow comments	8 years ago
dangqingqing	4eb25b4793	update code.	8 years ago
Yu Yang	133541ee41	Merge codes	8 years ago
Yu Yang	fbcadb75c6	Merge branch 'develop' of github.com:baidu/Paddle into gradient_check_utils	8 years ago
fengjiayi	589c3d864b	Rewirte test based on python test frameworks	8 years ago
Qiao Longfei	cc6c33b8b2	export Backward to python (#3174) ... * export Backward to python	8 years ago
dangqingqing	bd01cea1f3	Merge branch 'develop' of https://github.com/PaddlePaddle/Paddle into sigmoid_bp	8 years ago
dangqingqing	0560733c2e	Add sigmoid backward implenmention.	8 years ago
Yu Yang	3e812383bc	Merge branch 'develop' of github.com:baidu/Paddle into gradient_check_utils	8 years ago
Yu Yang	a0590004ce	Add __all__ in `gradient_checker.py`	8 years ago
qijun	341d1880ee	Merge remote-tracking branch 'baidu/develop' into op_gpu_test	8 years ago
Yu Yang	fcc28ccea2	Add comments	8 years ago
Yu Yang	7c42aad412	Initialize Gradient Checker ... Add get_numeric_gradient API and its unittest.	8 years ago
qijun	81cc7a3227	merge baidu/develop	8 years ago
fengjiayi	244c343b66	Fix a test error	8 years ago
fengjiayi	3a8017d148	Merge branch 'backward_dev' into dev_add_FillZerosLikeOp_test	8 years ago
fengjiayi	c59fe7a039	Add unittest for FillZerosLikeOp	8 years ago
fengjiayi	8bd73159c7	Add unittest for `FillZerosLikeOp`	8 years ago
liaogang	c36a093c5e	Fix conflict	8 years ago
liaogang	1e676f684d	Add mean op unit test in python	8 years ago
Yu Yang	5d134a03f1	Refine remove std::shared_ptr in Scope ... * Make interface of Operator to `const Scope&`	8 years ago
qijun	cf5ac5888e	reduce gpu memory allocation in op_test	8 years ago
qijun	61f94f0002	add EIGEN_USE_GPU macro to op.cu file	8 years ago
qijun	4a1f7bd21f	add gpu python op test	8 years ago
qijun	4cc42171db	merge baidu/develop	8 years ago
dongzhihong	933e55eb06	fix conflict	8 years ago
dongzhihong	47556689d4	"remove unused code"	8 years ago
dongzhihong	4d8ece8a3c	"update"	8 years ago
Yan Chunwei	aee0d3ec5f	RecurrentOp implementation (#2890) ... * add rnn op interfaces * add Run * rename state -> memory * change state -> memory * make compilable * add .cc * init test * add op fake implementation * add CreateStepNet and CreateScopes implementation. * add TODO list * init memory attributes. * add LinkMemories * add PlainNet fake implementation * Use std::shared_ptr<Scope> in the OpRunContext. * add test * disable mutable_data * finist segmentInput function * enable mutable_data with a trick * RNNOp test. * enable LinkMemories with mutable_data * update SegmentInput function with comments * finish ConcatOutput function * reformat inputs and attributes boot_memories * Refine unit test. * Refine unit test. * modify inlinks. * add OpDesc to Net * fix bug and update unit test. * move step scopes from inputs to outputs * fix merge conflict, update SegmentInput function * add RecurrentOpProtoAndCheckerMaker. * clean the codes * Abstract GetStepScopes and GetMaxSeqLen function * refine LinkMemories * Refine code and add some comments. * add backward core * update for develop branch. * add forward core * add forward algorithm * Add RecurrentGradientAlgorithm implenmention. * use CopyFrom and Slice function in RecurrentOp * add unit test for LinkMemories. * fix unit test. * use the latest tensor.h, solve conflict * add maker * move SegmentInput and ConcatOutput to details nameplace * unit test for RecurrentGradientAlgorithm. * apply OperatorBase * apply net operator. * move memorys to attributes * add RecurrentGradientOp * open test unit test in recurrent_network_op_test. * revert some files. * add RecurrentArgument and Link struct to simplify member variable. * rename. * move recurrent_op from framework to operators * add RecurrentGradientOp Init * fix name * fix Link.interal/external name * use namespace operators instead of framework * clean the code * use the latest add_op and mul_op, don't test backward now * Remove ScopePtr and OperatorPtr * add get_net to pybind * add test_recurrent_op.py * add random into gen_tensor * update to develop branch and refine some code. * add some comments.	8 years ago
dongzhihong	a22567ebef	"fix almost equal error"	8 years ago
dongzhihong	984225ecf1	"fix operator"	8 years ago
qijun	358261f0bd	fix gpu build error	8 years ago
Yu Yang	754f0c68da	Fix unittest	8 years ago
dongzhihong	32c15a2919	"random op test"	8 years ago
Yu Yang	b80590d70b	Remove test_plain_net	8 years ago
Yu Yang	0ceeacbe45	Make Scope can lookup variable name by variable ... * Refine unittest also	8 years ago
Yu Yang	0ab678e9e2	Add unittest for network	8 years ago
Yu Yang	c14f3e8ff2	Merge branch 'develop' into feature/middle_level_net_api	8 years ago
Yu Yang	9e4fac1ef4	Complete Add Op	8 years ago
Yu Yang	684563660b	Init commit	8 years ago
Yu Yang	1294b3c53e	Expose Net to Python (#2967) ... * Expose Net to Python * Expose PlainNet to Python, make python can add_op, complete_add_op * Provide a low level api to manipulate Net * Unittest for Net::DebugString	8 years ago
qijun	06acd6d0cb	add unittest for some basic OpKernels	8 years ago
Qiao Longfei	2685765905	add cross-entropy-op (#2965) ... * add cross-entropy-op * add infershape and compute * implement Infershape and compute of onehotcrossentropy op	8 years ago
Yu Yang	a98346f4cd	Add comment to `OpTestMeta` (#2968)	8 years ago
Qiao Longfei	e3b27d1998	Add sgd op (#2950) ... * a simplest SGD op	8 years ago
Yu Yang	21baa94b04	Merge branch 'feature/expose_net_op' into feature/middle_level_net_api	8 years ago
Yu Yang	0380bfb3cf	Expose Net to Python ... * Expose PlainNet to Python, make python can add_op, complete_add_op * Provide a low level api to manipulate Net * Unittest for Net::DebugString	8 years ago
Yu Yang	3402b6ad39	Add Unittest of add_two_op	8 years ago
Yu Yang	d3373c5b85	Skeleton Of fully connected operator ... * Fc operator is a grouped operator, which combined by may internal operators. * InferShape & Run a FC operator in Python.	8 years ago
Yu Yang	a89c7ffa94	Make Tensor <--> Numpy interactive in tensor.h ... * Follow review comments to seperate Tensor Numpy interactive methods in tensor.h. * Simplify logic for `CastToPyBufferImpl`, make it as one struct and in details namespace. * Remove `Scope` expose in Python, since it currently is useless. * Remove some debug functions.	8 years ago
Yu Yang	bde90be71b	Read/Write a Tensor Python ... Basically following http://pybind11.readthedocs.io/en/stable/advanced/pycpp/numpy.html * Use buffer protocol to return a view of Tensor. It can be cast to numpy array in Python. * Set a numpy array to a tensor.	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	a76f7ed2eb	Get OpProtos in Python ... * PyBind and SWIG of paddle cannot be load in a single Python process, lazy import all SWIG library of Paddle. Otherwise, the glog, gflags are imported twice in a same Python process. * Note that all PyBind11 return C++ std::string as an unicode. For protobuf, it is need be cast to `str` before use them. * Add unit test for Get `OpProtos`	8 years ago
Yu Yang	8da5587205	Init commit	8 years ago
Yu Yang	d027f47d7d	Default scope function ... `Paddle` manages Scope as programming language's scope. It just a thread-local stack of Scope. Top of that stack is current scope, the bottom of that stack is all scopes' parent. Invoking `create_var/get_var` can `create/get` variable in current scope. Invoking `enter_local_scope/leave_local_scope` can create or destroy local scope. A `scoped_function` will take a `function` as input. That function will be invoked in a new local scope.	8 years ago
Yu Yang	27b196ba6d	Expose paddle.framework by pybind11 (#2793) ... * Expose paddle.framework by pybind11 * Export paddle.framework.{Scope, Variable} to paddle.v2.framework.core. * See python/paddle/v2/framework/tests/test_scope.py for Python usage * See paddle/pybind/pybind.cc for C++ bind code. * add copyright	8 years ago
Yu Yang	e2ea1f42e9	Generate python protobufs for paddle.v2.framework ... Python should be able to manipulate Protobuf message because: 1. Python's `create_op_creation_methods` take the `OpProto` array to generate all `op_creation_methods` in RunTime. 2. All `op_creation_methods` will create an `OpDesc` and pass it to Paddle C++ method `CreateOp` and return the Op handle. Here is the list of what is added in this commit: * Add `protobuf_generate_python` if it is not defined. * Before cmake 3.4, `protobuf_generate_python` is not defined. Just copy the implementation of that function in `protobuf.cmake` * Add `py_proto_compile` function in `cmake/generic.cmake`. * It follows bazel's API interface. * https://github.com/pubref/rules_protobuf#rules * Add an empty package named `paddle.v2.framework`, all python code of `paddle::framework` will be in that package. * Generate protobuf's python module `__init__.py` by `touch` while compiling. * Change setup.py.in, make `paddle.v2.framework.proto` uses the generated protobuf pythons.	8 years ago