Paddle

Commit Graph

Author	SHA1	Message	Date
liaogang	90c4cd8323	FIX: header file	8 years ago
liaogang	2680dca9c8	ENH: add cuda enforce to PADDLE_ENFORCE	8 years ago
liaogang	f812de2cce	ENH: unify PADDLE_ENFORCE	8 years ago
gangliao	48cf64e833	Merge pull request #2674 from gangliao/cpu_mem [Done] Memory Management: Buddy Allocator	8 years ago
fengjiayi	c48fc4d826	Merge pull request #2825 from Canpio/dev_add_tensor_copy Add Tensor::CopyFrom and Tensor::mutable_data(Place place)	8 years ago
liaogang	03b3d0d8a8	Follow comments	8 years ago
fengjiayi	57a22db3fa	update PADDLE_ENFORCE message	8 years ago
fengjiayi	34beec0f7a	update tensor.h	8 years ago
fengjiayi	8594d5c314	change int numel_ to size_t numel	8 years ago
fengjiayi	8bcd1faffc	refactor product(DDim ddim)	8 years ago
Liu Yiqun	450cf18b53	Add Go compiler to Dockfile.android and rename the build directory to build-android. The newest developing image was push to dockerhub, named xreki/paddle-android:dev.	8 years ago
hedaoyuan	3f5e650d2d	Merge pull request #2862 from hedaoyuan/fix_14 Remove useless empty pointer check.	8 years ago
fengjiayi	1f97388a34	fix several compile error	8 years ago
fengjiayi	dcfcf6872a	Refactor Tensor::CopyFrom() 1. Add template T which indicates data type to `CopyFrom()`, `Slice()` and `ShareData()` functions. This makes `CopyData()` code much clearer. 2. Add `set_dim()`. 3. `product(DDim)` transforms `DDim` to `vector<int>` first and then calculate its product. That might be quite slow. For `product(dims_)` is frequently used in Tensor, we add a mumber variable `numel_` as a cache of the product result. TODO: refactor `product()` to make it more efficient. 4. Unable Tensor::operator= 5. Remove the limit of POD type, because `float16` and `int8` are not POD type.	8 years ago
fengjiayi	a1dc4311a3	Refactor `Tensor::CopyFrom()`	8 years ago
qijun	bac1426d47	add_op kernel implementation	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
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
hedaoyuan	010adb99b4	Remove useless empty pointer check.	8 years ago
liaogang	033523ea9d	update	8 years ago
qijun	6f2eba3e7e	Merge remote-tracking branch 'baidu/develop' into tensor_to_EigenTensor	8 years ago
liaogang	ea916c8487	Fix: alignment metric	8 years ago
Yu Yang	cd5113c197	Init commit	8 years ago
liaogang	21b7915d91	Fix condition compile	8 years ago
liaogang	365b457a04	Merge conflicts	8 years ago
liaogang	ab5fe1e907	ENH: memory test: check alignment and memory size	8 years ago
helinwang	f49fda5e73	Merge pull request #2855 from helinwang/issue/2854 Fix optimizer parameter buffer allocation size.	8 years ago
Yiqun Liu	bb8ac28b20	Merge pull request #2845 from Xreki/travis_build_android Add build_android task on Travis CI.	8 years ago
qijun	0b0b3ba16a	Merge remote-tracking branch 'baidu/develop' into tensor_to_EigenTensor	8 years ago
fengjiayi	dfa4650920	Merge branch 'develop' of https://github.com/PaddlePaddle/Paddle into dev_add_tensor_copy	8 years ago
qiaolongfei	e588730147	change op to operators	8 years ago
Helin Wang	11660eab0e	Fix optimizer parameter buffer allocation size. The buffer allocation size should be number of bytes, not number of floats.	8 years ago
Liu Yiqun	62908dcc24	Move the download of ndk to build_android.sh script file.	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
liaogang	00572aa451	Add memory alignment test	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
Liu Yiqun	95897fd1e1	Add build_android task on Travis CI.	8 years ago
qijun	f6fe37155f	Merge remote-tracking branch 'baidu/develop' into fix_bug_dynload	8 years ago
Qiao Longfei	728665d709	Add Init to OperatorBase (#2838 )	8 years ago
liaogang	ff98e3c1ec	ENH: Remove comments	8 years ago
qijun	71e2a94310	Merge remote-tracking branch 'baidu/develop' into tensor_to_EigenTensor	8 years ago
qijun	4e918377d0	fix bug in dynload	8 years ago
helinwang	f5f7d6bd4f	Merge pull request #2811 from helinwang/go_test_1 Add go testing into cmake	8 years ago
Helin Wang	e4be077ffa	Add go testing into cmake and fix libpaddle_go_optimizer.a link path	8 years ago
Qiao Longfei	be441f7d16	test OpKernel (#2820 ) Add unit test for OpKernel	8 years ago
hedaoyuan	891e5dcc48	Modify the default value of nnpack_allocate_outside.	8 years ago
Yu Yang	555b0a72d6	Merge pull request #2814 from QiJune/feature/refine_device_context Add GetEigenDevice To DeviceContext.	8 years ago
qijun	70d937c595	add memory header file	8 years ago
hedaoyuan	06748210d4	Fix some link errors about NNPACK.	8 years ago
fengjiayi	69d99d481d	Add Tensor::CopyFrom and Tensor::mutable_data(Place place) 1. Add `Tensor::CopyFrom`. Current version can only support CPU memory copy. The support of GPU will be provided later by `paddle::memory`. The current implementation of `Tensor::CopyFrom` is a little inefficient: Every time `CopyFrom` is called, tensor will re-allocate its memory. However, if we try to check and reuse `placeholder_`, we have to provide a template parameter for `CopyFrom` to indicate the data type. It seems strange for a simple copy function. 2. Add `Tensor::mutable_data(Place place)`, which directly use member variable `dims_` as its dim parameter. This interface is required by `Op::InferShape`.	8 years ago
qijun	be2c1a3b99	follow comments	8 years ago
qijun	a07deac9ef	follow comments	8 years ago
qijun	85806e7585	follow comments	8 years ago
qijun	8ee50a35d4	fix gpu build error	8 years ago
Luo Tao	e0ea87c99d	fix pybind compile question	8 years ago
qijun	ca23d86160	merge baidu/develop	8 years ago
qijun	4d336d9063	follow comments	8 years ago
Yu Yang	0a32008126	Merge pull request #2815 from reyoung/feature/op_kernel OperatorWithKernel	8 years ago
Yu Yang	0ff8192072	Add OperatorWithKernel class * User can register OpKernel to its Ops. The OpKernelMap saved in OperatorWithKernel. Each Op which inherits OperatorWithKernel will use `OpKernel::Compute` instead of Run.	8 years ago
qijun	ef5f9debc6	refine device_context	8 years ago
qijun	6bbc2944ae	fix code style	8 years ago
qijun	37bcc0860a	merge baidu/develop	8 years ago
fengjiayi	8681c5800b	Merge pull request #2813 from Canpio/dev_update_dim Remove Dim::contiguous and Dim::contiguous_strides	8 years ago
fengjiayi	76b4dd6226	Merge pull request #2809 from Canpio/dev_add_tensor_interface add tensor interfaces	8 years ago
qijun	b5a8d5b4b4	remove unused deps	8 years ago
qijun	8f5a9fd9a7	fix gpu build error	8 years ago
fengjiayi	2dbe60e489	Remove Dim::contiguous and Dim::contiguous_strides Paddle's data block is row-major order, while Dim::contiguous and Dim::contiguous_strides are based on column-order. So remove them to prevent misuse.	8 years ago
fengjiayi	2dccab87d4	fix some syntax problems	8 years ago
qijun	14d2c3990f	split device_context	8 years ago
QI JUN	2749b71f27	Merge pull request #2806 from reyoung/feature/refine_cuda_h Refine CUDA Related libraries	8 years ago
武毅	bcf9f421c3	Merge pull request #2774 from typhoonzero/fix_newupdater Fix new remote updater for go pserver	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
fengjiayi	0665dc9755	add more test	8 years ago
fengjiayi	ca39515e24	Add several interfaces for Tensor class 1. Add member variable 'DDim dims_' and a getter function 'dims()'. 'dims' is supposed to hold tensor's shape during Op::InferShape. 2. Remove 'mutable_data' which use default Place. User must specify a explicit Place when call 'mutable_data'. 3. A PlaceHolder may be shared by more than one tensor, and some of them may be the others' slices. So we add a new member variable 'offset_' for Tensor, which is used to show the byte offset between PlaceHolder::ptr_ and where tensor's data really begins. 4. Add functions 'ShareDataFrom' and 'Slice' for Tensor. TODO: Tensor needs a 'CopyFrom' function.	8 years ago
qijun	69d76812ae	fix cublas dynload bug	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
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
liaogang	383b96f32c	FIX: merge conflicts	8 years ago
qijun	a5eb1d8fab	fix build error	8 years ago
Yu Yang	a046605307	Refine CUDA Related libraries	8 years ago
liaogang	f404282dc2	Merge branch 'develop' of https://github.com/PaddlePaddle/Paddle into cpu_mem	8 years ago
liaogang	6a3b8416df	FIX: clang-format	8 years ago
liaogang	d4017cadcd	ENH: Add auto-free if allocate too much	8 years ago
qijun	d607f0b703	use cached rank	8 years ago
Yu Yang	abff52abcb	Simplelize framework/CMakeLists.txt (#2803 ) * generic.cmake can propogate dependencies through libraries. It is no need to specific all dependencies.	8 years ago
qijun	958511160b	add simple add_op_functor	8 years ago
liaogang	49fd49f76b	Fix conflicts	8 years ago
qijun	d6f7c3535d	move unaligned tensor types	8 years ago
dongzhihong	b871641a53	"switch to shared_ptr"	8 years ago
dongzhihong	18e65b0c08	"support net_proto header"	8 years ago
wanghaoshuang	de5ded6bbd	Merge branch 'develop' of https://github.com/PaddlePaddle/Paddle into crop_layer	8 years ago
yi.wu	5a4f33df7e	Merge branch 'develop' of https://github.com/PaddlePaddle/Paddle into fix_newupdater	8 years ago
dongzhihong	bc021d775e	"move opContext to DeviceContext"	8 years ago
qijun	1a968b4f64	init	8 years ago
dongzhihong	052d1d16ee	Merge remote-tracking branch 'origin/develop' into net_ycw	8 years ago
QI JUN	1038bc461a	implement DeviceContext (#2709 ) * add device_context * add unittest for device_context * transfer to use function paddle::platform::throw_on_error * fix cuda build error * using dynload functions * follow comments	8 years ago
liaogang	199b5fcb45	ENH: refine code comments	8 years ago
liaogang	1ce2fca484	Merge conflicts	8 years ago
dongzhihong	45adbfc4e0	Merge remote-tracking branch 'origin/develop' into lr_state	8 years ago
wuyi05	26d95a6bbf	fix new remote updater for go pserver	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
liaogang	ddfa6cf0d1	FIX: remove boost from memory folder	8 years ago
武毅	77604bca27	Merge pull request #2748 from typhoonzero/add_golang_precommit Add golang precommit	8 years ago
wanghaoshuang	0b788ef75b	Merge branch 'develop' of https://github.com/PaddlePaddle/Paddle into crop_layer	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
wuyi05	024a6537da	Merge branch 'develop' of https://github.com/PaddlePaddle/Paddle into add_golang_precommit	8 years ago
Yu Yang	1b366dc2ff	Fix CI error on test_LayerGrad.LSTM * We should not EXPECT_EQ between a float value and a int value. Use ASSERT_NEAR instead.	8 years ago
liaogang	a669bf48d9	FIX: explicit construct pool element	8 years ago
Superjom	1264480b04	fix ci	8 years ago
liaogang	5d2e8edb3e	FIX: dynamic loader deps	8 years ago
liaogang	936cd1e118	FIX: code format	8 years ago
liaogang	74691789e9	ENH: add memory unit test	8 years ago
yi.wu	9f64332258	Merge branch 'develop' of https://github.com/PaddlePaddle/Paddle into add_golang_precommit	8 years ago
Superjom	0b6430a363	Merge branch 'develop' of github.com:PaddlePaddle/Paddle into network	8 years ago
wanghaoshuang	d378e0a0cd	Merge branch 'develop' of https://github.com/PaddlePaddle/Paddle into crop_layer	8 years ago
Yang yaming	f21e3f731f	Merge pull request #2247 from pkuyym/fix-2240 fix bugs for CrossChannelNormLayer	8 years ago
liaogang	ada1c20bbc	FIX: Buddy Allocator Free with Merge feature	8 years ago
gangliao	1cc8fe7229	Merge pull request #2745 from gongweibao/fixcmake fix cmake errors	8 years ago
yangyaming	c37da0bd3b	Remove hasParamInitialValue flag.	8 years ago
Yu Yang	5d2a08775a	Merge pull request #2747 from reyoung/feature/correct_glog_check Correct GLOG CHECK in Paddle	8 years ago
gongweibao	204869c2da	fix bugs	8 years ago
gongweibao	2e302085d7	fix bugs	8 years ago
Tao Luo	7f380c1be3	Merge pull request #2701 from luotao1/stride stride pooling for max and average layer	8 years ago
Yu Yang	5eb8bf0324	Correct GLOG CHECK in Paddle Use CHECK instead of PCHECK, because PCHECK is used for errno.	8 years ago
liaogang	d0ad0314bb	FIX: glog dependency	8 years ago
wuyi05	7c6aa04f61	add go pre-commit and travis build	8 years ago
yangyaming	2bf4f1bbc1	make adding eps more clear	8 years ago
yangyaming	7ed6463ee9	fix bugs for CrossChannelNormLayer	8 years ago
Luo Tao	e7b071f33a	update annotation with comments	8 years ago
Superjom	309b37e52e	Merge branch 'develop' of github.com:PaddlePaddle/Paddle into network	8 years ago
wanghaoshuang	acfd2fc6df	fix cpp format	8 years ago
Yu Yang	9837896827	Merge pull request #2740 from reyoung/feature/remove_buggy_barrier_stats Remove buggy BarrierStat	8 years ago
wanghaoshuang	cf86891801	fix unittest of crop layer	8 years ago
wanghaoshuang	86bdb2f33f	fix crop function test	8 years ago
wanghaoshuang	5e6e1f636a	Add grad test and python wrapper for crop layer	8 years ago
wanghaoshuang	d1d70ec831	Refine configure option of crop layer 1. change configure content to 'axis, offset, shape' 2. add an optional input to crop layer as cropping reference	8 years ago
wanghaoshuang	e10040ca8a	add crop layer	8 years ago
Superjom	5cf9a23f9b	Merge branch 'develop' of github.com:PaddlePaddle/Paddle into network	8 years ago
Superjom	568c03ba1d	add virtual implementation	8 years ago
fengjiayi	830877f59b	Merge pull request #2706 from Canpio/dev_tensor_test fix compile errors in tensor. `PlaceholderImpl::Deleter` maybe used in other modules. We can consider moving it to be `paddle::memory::Deleter`.	8 years ago
Superjom	5c10a5ad55	remove virtual	8 years ago
Superjom	9f2357561d	fix ci error	8 years ago
Superjom	d1945e8d56	Merge branch 'develop' of github.com:PaddlePaddle/Paddle into network	8 years ago
helinwang	fa21616532	Merge pull request #2735 from gongweibao/fixcmake fix cmake error	8 years ago
liaogang	4dc3c9e0cc	ENH: Add paddle_memory for external usage	8 years ago
Yang yaming	312ce8b419	Merge pull request #2497 from pkuyym/ssd_outloss Add DetectionOutputLayer and MultiBoxLossLayer.	8 years ago
liaogang	0ba6347565	ENH: Add buddy allocator Free	8 years ago
Yu Yang	1ecddd8174	Remove buggy BarrierStat The implementation of BarrierStat is buggy, and it is not necessary for Paddle to diagnose which node in cluster is slow.	8 years ago
Qiao Longfei	571714159a	add op_desc.proto (#2736 ) * add op_desc.proto In Operator design, we need a proto message to describe an Operator. Third-party language such as python can build this proto message and use AddOp(const OpDesc& op_desc) of Paddle core to construct an Op in the Network.	8 years ago

1 2 3 4 5 ...

1966 Commits (0f4c7332969bdb057f855cd4a37174f3c06de281)