af4dac4ac3
Feature/free kid scope ( #4951 )
...
* Delete kid
* Delete local scope
7 years ago
4f41eaf7f8
Merge remote-tracking branch 'upstream/develop' into merge_grad_gtest
7 years ago
d75b00c221
refine the gtest log info and vlog order,
...
and change the size of test to make unit test faster
refine comment and log of mkldnnlayer
7 years ago
7eeaae1695
deconv
7 years ago
42f2dd4041
Unify `set_feed_variable` to one method ( #4949 )
7 years ago
c532b96741
Remove template parameter for Tensor methods ( #4937 )
...
* Remove template parameter for Tensor methods
* Also check the type is correct when data()
* Simplize holder_
* Fix accuracy_op
* Register Code
7 years ago
9e6404441c
fix elementwise add bug
7 years ago
d97a732f4f
deconv
7 years ago
e59ca752fc
Merge branch 'develop' of https://github.com/PaddlePaddle/Paddle into develop
7 years ago
77cac5cdb8
Removing updates of Beta1 power accumulators outside the op ( #4931 )
7 years ago
11bebeb2dc
Removing updates of Beta1 and Beta2 power accumulators outside the op ( #4925 )
7 years ago
3db5278301
Feature/py executor test ( #4922 )
...
* Implement FC layer with helper
* Update LayerHelper
* Add debug string for Python ProtoBuf
and Rename `Sync` to `Flush`
* Add check of ProtoBuf initialization
* Layer wrapper for FC
* Fix unittest
* Fix CI
* Add code generator
* AttributeChecker Better error log and speicalize bool
Since lots of types can be cast to bool
* Complete mlp, fit_a_line
* Expose get global scope
* Make global scope not thread-safe
1. It is no need to make global scope thread-safe, since it will be
invoked in Python main thread.
2. Do not free the global scope when C++ exit. Let the OS free memories,
otherwise, we need to handle the destroy dependencies.
See
https://google.github.io/styleguide/cppguide.html#Static_and_Global_Variables
* Fix
* Implementation of simple conv_2d layer
* Stash
* Remove private data members in OpRegister
* Fix bugs
* Stash
* Expose FeedFetchList as VarType
* Change ProgramDesc not a global variable
* Polish code style
* Stash
* Correct implement BlockDesc destructor
* Correct implement BlockDesc destructor
* Unify program as parameter name
* Fix bugs
* Add unittest
* Fix unit test error
* Remove unused functions
* Add clone for Python Program
* Working on executor
* Stash
* Add glog as dependencies of ops
* Use VLOG to logging some information is helpful when we debug Paddle
* Expose VarDesc::persistable to Python
* Test executor
* Complete unittest
* Polish code
* Fix merge error
* Follow comment
* Polish Python Code
7 years ago
56d5db8bea
Bug fix of libpaddle_capi_whole.a in x86.
7 years ago
747b541957
Follow comments
7 years ago
694bc64aaf
Merge branch 'develop' of https://github.com/PaddlePaddle/Paddle into lstm
7 years ago
17e33738f2
Enhance unit testing and fix bug.
7 years ago
63ffe5250a
Cluster train doc for v2 API ( #2072 )
...
* update cluster train v2 doc
* WIP cluster train doc
* update
* cluster train doc
* add TOC for en doc
* fix sphix build issue
* fix error links
* fix link errors
* fix image link
* polish cluster train docs
* update general distributed training document
* fix sphinx compile error
* fix doc image error
7 years ago
2073fb96cb
Enable learning rate annealing of Adam Optimizer
7 years ago
378dcb1833
Split paddle_capi_whole into paddle_nn_engine and paddle_layers two static libraries.
7 years ago
8bec26be03
Add missing file of math/detail/CMakeLists.txt
7 years ago
ee39b37a41
Merge branch 'develop' of https://github.com/PaddlePaddle/Paddle into lstm
7 years ago
74b283c9d6
Merge branch 'develop' of https://github.com/PaddlePaddle/Paddle into seq_expand_op
7 years ago
a94b3dd9a7
Refine comments and function name
...
1. Add more comments and exmples
2. Rename repeat_lod to expand_lod
3. Remove unused head file
7 years ago
94e442d4b1
add cpp file of MKLDNNLayer
7 years ago
9e38dafa29
change MKLDNNMatrix create interface since MatrixPtr is not always required
7 years ago
c1914543b0
refine mkldnn logic, move reset buffers into MKLDNNLayer
7 years ago
a461bf139d
Add missing file.
7 years ago
c8c4ba0f65
Merge pull request #4918 from wanghaoshuang/fix_lod
...
Make LoDTensor::lod_element return start and end offset of sequence
7 years ago
e8cd4b7d86
deconv2d impl in full
7 years ago
b325213150
update to develop branch.
7 years ago
d1fbf50b9e
Add unit testing for forwad implementation.
7 years ago
edb6aba698
make lod_element return std::pair
7 years ago
43aad989bd
deconv
7 years ago
47f773ddb2
Copy Constructor for ProgramDesc ( #4895 )
...
* Implement FC layer with helper
* Update LayerHelper
* Add debug string for Python ProtoBuf
and Rename `Sync` to `Flush`
* Add check of ProtoBuf initialization
* Layer wrapper for FC
* Fix unittest
* Fix CI
* Add code generator
* AttributeChecker Better error log and speicalize bool
Since lots of types can be cast to bool
* Complete mlp, fit_a_line
* Implementation of simple conv_2d layer
* Fix bugs
* Change ProgramDesc not a global variable
* Polish code style
* Stash
* Correct implement BlockDesc destructor
* Correct implement BlockDesc destructor
* Unify program as parameter name
* Fix bugs
* Add unittest
* Fix unit test error
* Remove unused functions
* Add clone for Python Program
* Compare OpDescBind directly
7 years ago
4018754d66
fix LoDTensor::lod_element to get last element in level
7 years ago
831927d58c
Merge pull request #4738 from tonyyang-svail/prune_impl
...
Prune implementation
7 years ago
e9249d16cb
Add glog as dependencies of ops ( #4908 )
...
* Add glog as dependencies of ops
* Use VLOG to logging some information is helpful when we debug Paddle
* Fix Unittests
7 years ago
f6e1d959d2
Expose VarDesc::persistable to Python ( #4911 )
7 years ago
fdfc8f9baa
"switch to Init op"
7 years ago
7ad13fbfe9
Merge pull request #4876 from QiJune/sgd_op_sparse_kernel
...
add sparse update kernel for sgd operator
7 years ago
5ec55e7995
deconv impl
7 years ago
c5b411c515
make compatible to new programDescBind
7 years ago
39aa81e74e
Merge remote-tracking branch 'upstream/develop' into prune_impl
7 years ago
80ebc8d5e3
Merge branch 'develop' of https://github.com/PaddlePaddle/Paddle into develop
7 years ago
c10b8e808f
Adding Proximal Gradient Descent ( #4848 )
...
* Adding Proximal Gradient Descent
* Fixing review comments
7 years ago
a204fefe16
Fix several bugs in compile time backward and Protobuf desc ( #4894 )
...
* Implement FC layer with helper
* Update LayerHelper
* Add debug string for Python ProtoBuf
and Rename `Sync` to `Flush`
* Add check of ProtoBuf initialization
* Layer wrapper for FC
* Fix unittest
* Fix CI
* Add code generator
* AttributeChecker Better error log and speicalize bool
Since lots of types can be cast to bool
* Complete mlp, fit_a_line
* Implementation of simple conv_2d layer
* Fix bugs
* Correct implement BlockDesc destructor
* Fix bugs
* Fix unit test error
* Follow comments
7 years ago
af215a1a53
Design doc: Batch Normalization Operator ( #3748 )
...
* Add design doc of batch_norm_op
* Move batch_norm_op.png to operator/images
* Refine batch_norm_op design doc
7 years ago
e747623e86
Change ProgramDesc not a global variable ( #4879 )
...
* Change ProgramDesc not a global variable
* Polish code style
* Correct implement BlockDesc destructor
* Unify program as parameter name
7 years ago
31531ab581
Add backward kernel
7 years ago
91db457fc0
follow comments
7 years ago
8de04be786
Fix unitest
7 years ago
2a8dbd130d
LSTM Operator forward implementation.
7 years ago
40688d223e
refine im2col (up_pad,down_pad)
7 years ago
d2c1408f91
fix im2col kocf for sequence projection
7 years ago
1e60c9b2e8
Add sequence_project_op (use im2col)
7 years ago
555ab3f5af
Merge branch 'develop' of https://github.com/PaddlePaddle/Paddle into seq_expand_op
7 years ago
23701ffaf0
Refine op
7 years ago
43c05e434d
Merge pull request #4855 from tonyyang-svail/executor_refine
...
Simplify executor
7 years ago
8938a9b0ac
Correct implement BlockDesc destructor ( #4882 )
7 years ago
521514da36
fix conflict ( #4883 )
7 years ago
164898277c
MatMul operator ( #4856 )
...
* initial matmul operator
Similar to np.matmul, but also has transpose_X and transpose_Y flags,
and only supports tensors from rank 1 to 3 inclusive.
For GPU, uses cublas?gemmStridedBatched. For CPU, uses
cblas_?gemm_batch if available via MKL; otherwise a simple serial
implementation that loops over the batch dimension is employed for now.
7 years ago
fd96914d23
Merge pull request #4571 from sidgoyal78/momentum_op
...
Add momentum operator
7 years ago
f9681459b2
fix gpu build error
7 years ago
5d67677c60
Remove private data members in OpRegister ( #4871 )
7 years ago
ab8cc401e6
add sparse sgd operator unittest
7 years ago
330c509bc3
Merge remote-tracking branch 'baidu/develop' into sgd_op_sparse_kernel
7 years ago
182ce51c6d
add sparse kernel of sgd operator
7 years ago
98dccc9f21
Merge branch 'develop' of https://github.com/PaddlePaddle/Paddle into develop
7 years ago
46f7854e5c
fix mac compile ( #4870 )
7 years ago
df0946ebe2
Impl optimizer ( #4734 )
...
* init parameter base class
* optimize the Comments of optimizer
* basic implimentation of optimizer
* add test_optimizer
* add no_grad_set to interface
* update optimizer.py
* python code can run
* fix some problem
* add sync_with_cpp to Python Program and Block
* sync vars and ops in block from cpp
* optimize code and add some comment
* add more check for sync
* update optimizer with return value of Backward
* rm unused code
* infer shape when create gradient vairiable
* update test_optimizer
* update test_program.py
* update backward test
* follow comment
7 years ago
451863dba2
Merge branch 'develop' of https://github.com/PaddlePaddle/Paddle into develop
7 years ago
426f7eee8e
simplify test_pool_py, add comments for different pooling strategy
7 years ago
60b84856e7
refine the conf files, combine in one file
7 years ago
2c1b35caaa
Merge branch 'develop' into maxseq
7 years ago
a8a63d4c50
add MAX strategy for seqpool op
7 years ago
3123e3cf5d
Merge branch 'develop' into crf
7 years ago
b487c62aac
Merge remote-tracking branch 'upstream/develop' into merge_grad_gtest
7 years ago
2dc6cea65c
add unit tests for mkldnn branches, and fix typo
7 years ago
064c3695ec
Make global scope not thread-safe ( #4850 )
...
* Make global scope not thread-safe
1. It is no need to make global scope thread-safe, since it will be
invoked in Python main thread.
2. Do not free the global scope when C++ exit. Let the OS free memories,
otherwise, we need to handle the destroy dependencies.
See
https://google.github.io/styleguide/cppguide.html#Static_and_Global_Variables
* Revert "FIX: Release CPU/GPU memory via deleter"
This reverts commit 8f80f5bc79
7 years ago
bdca4b37c4
change api based on design doc
7 years ago
e0cee58c84
modify protobuf
7 years ago
cf743ab064
merge develop
7 years ago
73a8b78a72
Correct OpWithKernel's infershape ( #4847 )
...
They are public now
7 years ago
cc220eec36
add forward computation of crf operator.
7 years ago
fafc54d6c0
fix var_type_inference_test ( #4854 )
7 years ago
f06637002d
simplify executor; pass compile
7 years ago
172e460d50
Merge pull request #4797 from reyoung/feature/implenment_infer_var_type
...
Complete infer_var_type
7 years ago
b10cd43554
rm cpp executor_test, rewrite in python later ( #4849 )
...
* rm cpp executor_test, rewrite in python later
* remove executor_test code in CMakeList.txt
7 years ago
cdc236cb82
Merge pull request #4841 from QiJune/pybind_selected_rows
...
export SelectedRows to Python
7 years ago
c65bdd9565
fix SelectedRows rows() method gpu runtime error
7 years ago
9fc593603b
Merge pull request #4837 from QiJune/pybind_feed_fetch_method
...
export feed/fetch variable method to Python
7 years ago
75d0c79015
Change Name convention of operator attributes ( #4807 )
...
* Change dataType to data_type
Follow PEP8
* Change name_convention to fit PEP8
7 years ago
186d165565
AttributeChecker Better error log and speicalize bool ( #4840 )
...
* AttributeChecker Better error log and speicalize bool
Since lots of types can be cast to bool
* add FIXME comment
7 years ago
97069927c9
Merge remote-tracking branch 'baidu/develop' into pybind_selected_rows
7 years ago
259f07b33c
Merge pull request #4844 from QiJune/remove_OpRegistrar
...
remove unused C++ class OpRegistrar
7 years ago
86acf39cce
Merge pull request #4801 from QiJune/add_selected_rows_functor
...
add some basic math functor for SelectedRows
7 years ago
eb27c735cb
remove unused C++ class OpRegistrar
7 years ago
865c2c8ed8
add compile DEPS
7 years ago
219f46ae16
export SelectedRows to Python
7 years ago
a64a6f527b
id to block_id
7 years ago
23cb8259c3
"add python test case"
7 years ago
5fe3f8f630
refine code
7 years ago
701c90f6f7
remove unused header file
7 years ago
7c48335b7c
merge fix linking
7 years ago
c87e060c18
export feed/fetch method to Python
7 years ago
eb187366f4
merge develop
7 years ago
517e3c7947
Merge remote-tracking branch 'baidu/develop' into pybind_feed_fetch_method
7 years ago
5993497cb7
fix merge conflict
7 years ago
4df6cf4d16
Rewrite feed/fetch op ( #4815 )
...
* Feed/Fetch op just plain operator, not a OpWithKernel
* Do not register OpInfoMaker since Feed/Fetch will never be
configured by users
* Feed/Fetch op has empty gradient
* Feed/Fetch op do not hard code `feed_variable`, `fetch_variable` as
its input and output, make it as a plain Operator input/output
7 years ago
571f43ca39
Merge remote-tracking branch 'baidu/develop' into add_selected_rows_functor
7 years ago
1199aa6876
fix bug: clear grad and always share data when output has cpu
...
and add activation in unit tests
7 years ago
6a4282a20f
refine comments of sequence_pool_op
7 years ago
216b81ace7
Merge branch 'develop' into seqpool
7 years ago
4da6e86ff7
make listen queue configurable ( #4659 )
7 years ago
0c72649afc
Fix gometalinter versioning ( #4832 )
...
* fix gometalinter versioning
* stop gometalinter
7 years ago
0d7b1fd4fe
fix bug in SelectedRowsAddTensorKernel
7 years ago
ab5dc9fe18
remove SelectedRows functors to selected_rows_functor.h
7 years ago
97332d7c0c
Merge branch 'develop' into seqpool
7 years ago
dee7f81325
add finish work of mkldnn
7 years ago
83627d3efd
Merge branch 'develop' of https://github.com/PaddlePaddle/Paddle into lstm
7 years ago
3cace73701
Add lstm implementation.
7 years ago
7c0facd195
init
7 years ago
60c2536d22
Merge branch 'develop' of https://github.com/PaddlePaddle/Paddle into develop
7 years ago
44ed21eea5
Refine backward unit tests ( #4817 )
7 years ago
73883bde2a
"fix error"
7 years ago
d4d215a514
Merge remote-tracking branch 'origin/develop' into multigpu
7 years ago
5bcb63800e
"fix common test"
7 years ago
d8aebaf50c
"fix enforce error"
7 years ago
6729f32c74
Expose Executor to Python ( #4804 )
...
* Expose Executor to Python
* Follow comments
7 years ago
a4b172252b
Merge pull request #4809 from jacquesqiao/backward-return-map
...
Backward return map
7 years ago
17b4cea49e
Merge pull request #4759 from tensor-tang/merge_grad
...
Merge mkldnn output grad
7 years ago
2befb9f972
optimizer backward CreateGradVarInBlock input output order
7 years ago
54793e3588
Merge remote-tracking branch 'origin/develop' into fix/scope
7 years ago
9a0ef7d2aa
append_backward return map to python
7 years ago
d2be7ec370
Merge remote-tracking branch 'origin/develop' into multigpu
7 years ago
89758adb83
fix code style
7 years ago
ec783d6b58
Feature/backward return map ( #4806 )
...
* Final step of backward, return a map from param_name to grad
* Complete the final step of backward
Return the param_name to grad_info
7 years ago
7ef568e893
fix gpu unittest error
7 years ago
5bd14f6edb
Merge remote-tracking branch 'origin/develop' into fix/scope
7 years ago
d7383c6dd0
create grad_var when run Backward pass ( #4796 )
...
* add target to Backward, generate var in block when call backward
* modify backward_test
* fix executor_test
* set var desc default type to LOD_TENSOR
* update backward_test
* insert loss in the top level of backward
* create grad vars for all blocks in current program
* optimize code
* update test_program.py
* only create var for newly create blocks when backward
7 years ago
d4231b3513
"fix unique_ptr error"
7 years ago
4130e5faf8
Merge remote-tracking branch 'baidu/develop' into add_selected_rows_functor
7 years ago
f59a7c1d36
add gpu functor for SelectedRows
7 years ago
ec11514027
"fix findVar logic"
7 years ago
29819ba764
Fix unittest
7 years ago
acc54c7ba7
Merge branch 'develop' of github.com:baidu/Paddle into feature/implenment_infer_var_type
7 years ago
b3df1f4a49
"fix tests"
7 years ago
2434b8f5fb
" add interface to scopeDesc bind"
7 years ago
3f4177ea8d
Merge branch 'develop' into fix/scope
7 years ago
54d3dbd8c9
"add enforce check"
7 years ago
24a796fbc0
Merge branch 'develop' of https://github.com/PaddlePaddle/Paddle into Add_conv3d_gemm_op
7 years ago
4741266d6f
remove unused method
7 years ago
931572e210
SelectedRowsAddTensor method
7 years ago
d144310415
"nccl add interface"
7 years ago
5d9ce04625
Debug string for Python ProtoBuf ( #4800 )
...
* Add debug string for Python ProtoBuf
and Rename `Sync` to `Flush`
* Add check of ProtoBuf initialization
7 years ago
2c46666e7b
Add grad_name_map to record correspondences between vars and grad_vars ( #4794 )
...
* Add grad_name_map
* Fix bug
* Fix bug
* Follow comments
7 years ago
7b18343305
remove unused header file
7 years ago
5be10872f9
add selected_rows add cpu functor
7 years ago
a96372b108
Merge branch 'develop' of github.com:baidu/Paddle into feature/implenment_infer_var_type
7 years ago
1b1cb44f13
Complete infer_var_type
7 years ago
652f182dc0
deconv
7 years ago
9165235a9e
Merge pull request #4782 from QiJune/selected_rows
...
add selected rows
7 years ago
da399ae6b1
Merge branch 'develop' of https://github.com/PaddlePaddle/Paddle into develop
7 years ago
d17eb73e9f
Update VarDesc from design doc ( #4769 )
...
* Update VarDesc from design doc
* Fix GCC compile
* Fix unittest
7 years ago
9a6dffd4d9
Fix CPU compile ( #4781 )
7 years ago
7a7c8fd9e5
simplify some comments and code
7 years ago
abfa81b106
Merge branch 'develop' into seqpool
7 years ago
918d69467c
Merge pull request #4791 from typhoonzero/fix_pybind_op_reg_nokernel
...
Fix pybind op register nokernel
7 years ago
a0af1eeabf
Merge pull request #4443 from guoshengCS/add-GRUStepOp
...
Add gru_unit_op
7 years ago
557c7ae37b
remove conflict
7 years ago
1540703394
fix_pybind_op_reg_nokernel
7 years ago
59ccb01a00
Merge remote-tracking branch 'upstream/develop' into merge_grad
7 years ago
13b4749e24
fix_pybind_op_reg_nokernel
7 years ago
e1954857d7
fix bug: merge grad must before backward act.
...
and add branch net comparing with cpu result
7 years ago
3b954e1ddc
Adding Hard Sigmoid Activation ( #4771 )
...
* Adding Hard Sigmoid Activation
* Adding a comment for slope to be only positive
* Fixing grammatical mistake in comment
7 years ago
1c1f73b46d
Feature/dynamic recurrent op forward test ( #4729 )
7 years ago
53542d93e5
Merge remote-tracking branch 'baidu/develop' into selected_rows
7 years ago
4c3ef7fca5
Merge branch 'develop' into seqpool
7 years ago
c49adb86c6
follow comments
7 years ago
11e923176e
reset a new tensor in default constructor of SelectedRows
7 years ago
4b13c80eeb
add selected rows
7 years ago
a36d24163a
Add no_grad_vars for grad_op_maker ( #4770 )
...
* Add no_grad_vars for grad_op_maker
* Add unittest
* Fix unittest
* Fix unittest
* Follow comment
7 years ago
4cda9a36a4
Stablize executor_test ( #4774 )
...
Use less GPU memory
7 years ago
4838ea25d3
Wrong dependency order for op_info and proto_desc ( #4763 )
7 years ago
416f590912
Merge branch 'develop' of https://github.com/PaddlePaddle/Paddle into develop
7 years ago
116800378a
Adding the Adam Optimizer operator ( #4733 )
...
* add adam op
moment1_out = beta1 * moment1 + (1 − beta1) * grad
moment2_out = beta2 * moment2 + (1 − beta2) * grad * grad
moment1_hat = moment1_out / (1 - beta1^t)
moment2_hat = moment2_out / (1 - beta2^t)
param_out = param - learning_rate * moment1_hat / (sqrt(moment2_hat) +
epsilon)
* fix moment 2
* Adding the Adam optimization operator
* Adding more tests for Adam op
7 years ago
7da9ab3c6c
Merge remote-tracking branch 'origin/develop' into fix/scope
7 years ago
7460958143
Merge pull request #4764 from jacquesqiao/remove-infer-todo
...
remove unused todo in shape_inference.h
7 years ago
be85516318
Remove attribute check from CreateGradOpDescs() ( #4723 )
...
* Remove attr_checker from CreateGradOpDescs()
* Fix merge error
* Fix bug in backward_test.cc
7 years ago
9ddedeee06
remove unused todo in shape_inference.h
7 years ago
ff0e9d2207
Merge remote-tracking branch 'origin/develop' into fix/scope
7 years ago
1a87a963e1
Merge branch 'develop' of https://github.com/PaddlePaddle/Paddle into dev_opdesc_in_python
7 years ago
dd4f95f6c4
Merge branch 'develop' of https://github.com/PaddlePaddle/Paddle into develop
7 years ago
7f0c823093
lambda cost bug's fix
7 years ago
698071cc33
share mkldnn output value data if next layer is cpu device
7 years ago
71abb4ceb1
Merge branch 'develop' of https://github.com/PaddlePaddle/paddle into add-NormOp
7 years ago
ae1b29ac7e
Merge branch 'develop' of https://github.com/PaddlePaddle/paddle into add-GRUStepOp
7 years ago
cbcf11d930
Merge branch 'develop' into crf
7 years ago
393c748c89
add seqlastin/seqfirstin for seq_pool op
7 years ago
4aae1fff78
fix conv3d_gemm, unit test and follow comments
7 years ago
d0d3129f91
Fix InferShapeContext in decayed_adagrad_op
7 years ago
9b4a6af251
Merge branch 'develop' of https://github.com/PaddlePaddle/paddle into add-GRUStepOp
7 years ago
0bc5a122d4
Refine gru_unit_op by optional bias
7 years ago
1cabdb8708
Refine gru_unit_op according to comments to support multiple activation types
7 years ago
d3b8bffaf1
Implementing the Decayed Adagrad optimizer operator ( #4645 )
...
* Implementing the DecayedAdagrad optimizer step operator
* implementing DecayedAdagrad operator
* remove file
* small fix
7 years ago
e69a565a5b
Merge branch 'develop' into seqpool
7 years ago
2daba04042
Merge pull request #4731 from reyoung/feature/fix_ci
...
Feature/fix ci
7 years ago
a3ccbdb3b6
Cudnn conv op ( #4195 )
...
* add cudnn_conv_op
* WIP
* update
* update
* fix grad check
* use platform::memory
* add support group for cudnn
* update
* follow comments
* fix onlycpu build
* update cuda define
* follow comments
* follow comments
* merge with updates
* fix compile error
* follow comments
* follow comments
7 years ago
4118782b31
Merge remote-tracking branch 'origin/develop' into multigpu
7 years ago
0fa34db759
nccl init
7 years ago
9106a4bba1
Merge branch 'develop' of https://github.com/PaddlePaddle/Paddle into lstm
7 years ago
8728b3cce2
Add LSTM Operators.
7 years ago
91cc5d6208
add the forward operator.
7 years ago
515981d714
Merge branch 'develop' into crf
7 years ago
d92c671d5f
add python forward unittest.
7 years ago
35308d3662
"change NewVar to Var"
7 years ago
6715beaace
enable merge output grad of mkldnn
7 years ago
58b8a1ae4c
prune link fail
7 years ago
4b6b4bc84a
"change GetOrCreate to Var"
7 years ago
1dd6dbbce2
deconv
7 years ago
fc96463b25
pass multiple target
7 years ago
fd72e9c751
pass multiple unit test
7 years ago
848febbe96
"fix typo"
7 years ago
a31ff363fd
prune pass dummy test
7 years ago
458a2da3cf
Merge branch 'fix_bugs' into dev_opdesc_in_python
7 years ago
2434e4862a
Fix bugs
7 years ago
eb33d53732
Merge remote-tracking branch 'pr/4730' into feature/fix_ci
7 years ago
f4ec8bdb4d
Merge remote-tracking branch 'pr/4721' into feature/fix_ci
7 years ago
f4b32673ca
correct op deps in executor_test
7 years ago
752bab2719
Fix bugs
7 years ago
65b8861cde
Try to fix error: cannot allocate memory in static TLS block
...
ImportError: /usr/local/lib/python2.7/dist-packages/py_paddle/_swig_paddle.so: cannot allocate memory in static TLS block
7 years ago
a4687c9909
fix conflict of CopyFrom in vol2ol_test
7 years ago
5d6a3eee5b
new_var to get_or_create
7 years ago
0af45b5fa6
NewVar to GetOrCreateVar
7 years ago
b504a2346c
Adding the Thresholded Relu Op ( #4685 )
...
* Adding thresholded_relu op
* Adding test for thresholded relu op
7 years ago
2603cb7e86
Unify CUDA stream in Tensor CopyFrom interface ( #4692 )
...
* init
* unify CopyFrom interface
* fix gpu build error
* fix bug in tensor_py.h
* refine code comments and add TODO list
* fix conflicts in FeedOp and FetchOp
7 years ago
8e7975da3f
move GLOB_OP_LIB deps to executor_test
7 years ago
bf26cc5307
Merge branch 'develop' of https://github.com/PaddlePaddle/Paddle into dev_opdesc_in_python
7 years ago
ccea4c5772
debug executor_test
7 years ago
d92f8de431
Merge pull request #4702 from Canpio/fix_expose_checker_to_python
...
Explose check_attr to Python
7 years ago
5e9d43995b
Merge pull request #4699 from jacquesqiao/expose-backward
...
expose AppendBackward of ProgramDesc to python
7 years ago
f984cba0ca
Merge branch 'develop' of https://github.com/PaddlePaddle/Paddle into seq_expand_op
7 years ago
acd1aaea49
fix issues
7 years ago
0922fca41e
Add gru_unit_op
7 years ago
901b041196
Add seq_expand op
...
1. Add unitest
2. Add SeqExpandOpKernel
7 years ago
e9495e7694
Merge pull request #4508 from Yancey1989/seqconcat_op
...
Add the sequence_concat operator.
7 years ago
4b1f70d99a
Merge pull request #4716 from luotao1/bug
...
fix executor_test build failed
7 years ago
b56cbd307d
Merge pull request #4285 from kuke/margin_rank_loss_op_dev
...
Add margin rank loss operator
7 years ago
d68122ff5d
update
7 years ago
69e92b39fc
add an enforce
7 years ago
1f592eb8b6
pause executor_test temporary in order to pass the teamcity
7 years ago
96b4035dd1
Add conv3d_gemm_op
7 years ago
98dd5b1ac8
fix executor_test build failed
7 years ago
c77bd553de
Merge branch 'develop' into seqpool
7 years ago
e9325ea8b1
Merge pull request #4676 from chengduoZH/fix_pool_op_doc_and_class_declarations
...
Fix pool op doc and class declarations
7 years ago
989e19caf9
fix typos in margin_rank_loss_op
7 years ago
13b7d92808
improve doc in margin_rank_loss_op
7 years ago
4849fba7e9
follow comments
7 years ago
240adef1a1
Merge branch 'develop' of upstream into margin_rank_loss_op_dev
7 years ago
2f17038428
Merge pull request #4707 from luotao1/bug
...
fix Compile error [fatal error: boost/range/adaptor/reversed.hpp]
7 years ago
ad477b9183
update
7 years ago
1cafe7bfa6
Merge pull request #4703 from Xreki/core_optimize_backward
...
Simplify backward when inserting a sum operator to accumulate all duplicated variables
7 years ago
82a2b1a92d
fix Compile error [fatal error: boost/range/adaptor/reversed.hpp No such file or directory]
7 years ago
c22e7ff71e
Merge branch 'develop' into fix_pool_op_doc_and_class_declarations
7 years ago
e8cad5a1d0
add more unit test for test_append_backward
7 years ago
62da438e8e
Merge pull request #4462 from chengduoZH/Add_vol2col_functor
...
Add vol2col functor
7 years ago
9995aed114
Implementing Softplus operator ( #4690 )
...
* implementing softplus
* small fix
* small fix
* small fix
* small fix
7 years ago
7454ec0400
Simplify backward when inserting a sum operator to accumulate all duplicated variables.
7 years ago
c3bf332666
Merge pull request #4537 from QiJune/executor_impl
...
Executor interface design and implementation
7 years ago
f8267db657
Explose check_attr to Python
7 years ago
2e554693cc
assgin todo to a certain person
7 years ago
f5ac335046
follow comments
7 years ago
e9a0c4ef87
expose AppendBackward of ProgramDesc to python
7 years ago
1397e17f6b
Implemented the hardShrink activation ( #4653 )
...
* Implemented the hardShrink activation
* Fixing the unit test
7 years ago
67dc606b32
Merge branch 'develop' into seqpool
7 years ago
1d41a6d45a
update paddle/operators/math/CMakeLists.txt
7 years ago
e85fa5e9c1
Merge branch 'develop' into fix_pool_op_doc_and_class_declarations
7 years ago
32146da9b1
Merge branch 'develop' of https://github.com/PaddlePaddle/Paddle into Add_vol2col_functor
7 years ago
c85d777f87
follow comments
7 years ago
72d3d814b5
fix math/CMakeLists.txt
7 years ago
6604d7cda2
Add logsigmoid (numerically stable) and softshrink ( #4663 )
...
* Add numerically-stable logsigmoid activation
* Add softshrink operator
* Adjust relative tolerance for grad-check
* Address review comments
7 years ago
e71b836f53
Merge branch 'develop' of https://github.com/PaddlePaddle/Paddle into dev_opdesc_in_python
7 years ago
247fb2a086
Add unittests
7 years ago
ee22a436a8
Merge pull request #4684 from reyoung/feature/parameter
...
Feature/parameter
7 years ago
843ed8e320
dynamic recurrent op forward c++ implentation ( #4597 )
7 years ago
434949ca2d
clean up for merge
7 years ago
f410622f03
merge follow comment
7 years ago
a528a9717e
remove prune as member function to function
7 years ago
436ea50d5f
follow comments
7 years ago
2e7cd201a4
remove log in backward
7 years ago
0c37a06117
Merge branch 'feature/change_proto_to_desc' into feature/complete_variable_bind
7 years ago
062ff4d77b
clean up
7 years ago
7506e4816b
Merge pull request #4660 from reyoung/feature/polish_infer_shape
...
Polish CompileTime InferShape
7 years ago
293a7d1e75
add feed infershape todo
7 years ago
3f9e247a73
set variable support dim
7 years ago
83dbc15055
Merge branch 'develop' of github.com:baidu/Paddle into feature/complete_variable_bind
7 years ago
805639b16c
Fix compile error in linux
7 years ago
a281b38393
Conv Shift Operator ( #4591 )
...
* conv_shift_op: initial implementation using Eigen
Limitations:
- both gradient outputs must be specified and are always computed
- explicit for loops => could be optimized in various ways
(e.g., different memory layout)
* conv shift - gradient fixes
fix case when not all output gradients desired
* conv shift: minor cleanup
* conv shift - more minor cleanup
* conv shift: clean up & initial GPU implementation
* fix rebase issue
7 years ago
69fd376bca
Merge branch 'develop' of https://github.com/PaddlePaddle/Paddle into feature/polish_infer_shape
7 years ago
a308ff29af
make infershape of feedop and fetchop compatible with compile-time design
7 years ago
cec1f598c6
Merge pull request #4664 from Canpio/fix_backward_bug
...
Fix OpDesc bug
7 years ago
0aada73af1
Merge pull request #4673 from luotao1/onlycpu
...
remove unused PADDLE_ONLY_CPU comments
7 years ago
f2e7cf2141
fix InferShapeContextBase to InferShapeContext
7 years ago
6db476ed89
Separate the declarations and implementation of the PoolOp and PoolMaker class in order to reuse in pool_cudnn
7 years ago
67edd04a2f
fix doc
7 years ago
0f1d3af438
Merge pull request #4461 from chengduoZH/Add_maxpool_withIdx_only
...
Add max pool op (with index)
7 years ago
60f706a1d6
add SQRT strategy for sequence_pool_op
7 years ago
36da82550a
Add code comments
7 years ago
871a3f6e76
remove unused PADDLE_ONLY_CPU comment
7 years ago
975a51294e
infer feed operator output variable shape with dims attribute
7 years ago
e21e5646a5
fix atomicAdd -> CudaAtomicAdd
7 years ago
e880a356fe
update
7 years ago
2fc7fc7a18
pass multiple forward backward
7 years ago
462579c416
update
7 years ago
8e2cc75435
Merge pull request #4667 from PaddlePaddle/reyoung-patch-1
...
Removed unreached code
7 years ago
e3161bb61a
pass simple backward
7 years ago
a4d410aec8
Merge branch 'develop' of github.com:PaddlePaddle/Paddle into seqconcat_op
7 years ago
d211b51bd4
update comment
7 years ago
a17442d4be
merge 4664 in advance
7 years ago
32cb74be3e
Removed unreached code
7 years ago
cb2ef7d97d
Add comment
7 years ago
0e1f21a570
Fix bug
7 years ago
7d21d8c022
Merge remote-tracking branch 'baidu/develop' into executor_impl
7 years ago
15400748ae
follow comments and refine codes
7 years ago
6c6474cbd8
follow coments
7 years ago
e6421249d5
update to latest
7 years ago
932402c16b
debug for sum
7 years ago
d350cdbc97
Polish CompileTime InferShape
...
1. InferShape should be a method for `OpDescBind` not `Operator`, since
`OpDescBind` is a compile-time concept.
2. Pre-create operators and store them into a map. Make InferShape
faster
7 years ago
bb81baa19b
Merge pull request #4658 from luotao1/bug
...
fix compiler error: ‘unique_ptr’ is not a member of ‘std’
7 years ago
9efd5422f9
Merge pull request #4655 from abhinavarora/fill_constant_op
...
Implementing the fill constant op for the executor
7 years ago
ceefb555f7
Merge pull request #4500 from luotao1/interp
...
add interpolation op
7 years ago
c876cacc7b
fix compiler error: ‘unique_ptr’ is not a member of ‘std’
7 years ago
6efacc14d8
Implementing the fill constant op for the executor
7 years ago
e655d29173
merge develop
7 years ago
340d21d4ed
Init at block[0]; Run at block[1]
7 years ago
b33779d653
Merge pull request #4648 from reyoung/feature/use_proto_lite_for_refacting
...
Use PROTO_LITE when refactoring Paddle
7 years ago
d2217ffa39
Fix CI
7 years ago
e51557130e
clean up for review
7 years ago
ce901b1186
Refine unit test
7 years ago
3a8de910f3
Merge branch 'develop' of https://github.com/PaddlePaddle/Paddle into dev_add_attr_test
7 years ago
5984cbca47
Add Attr test
7 years ago
383faaf719
Adding implementation for copying a vector to a tensor ( #4635 )
...
* Adding implementation for copying a vector to tensor
* Changing Tensor test to access gpu memory indirectly
7 years ago
dcb09e932d
Use PROTO_LITE when refactoring Paddle
...
It will significantly reduce binary size. It is useful for mobile
deployment.
7 years ago
d23cd51450
Merge pull request #4646 from reyoung/feature/fix_compile_error
...
Fix compile error in develop branch
7 years ago
4be2f44c68
Merge branch 'feature/fix_compile_error' into feature/complete_variable_bind
7 years ago
6600e2392f
Merge pull request #4644 from reyoung/feature/backward_default_values
...
Fix bug of foward default attribute not passed to backward
7 years ago
92add2a29b
Fix compile error in develop branch
7 years ago
49ca0b4831
Stash
7 years ago
e12ec95ac1
Merge pull request #4630 from jacquesqiao/merge-infershapecontext
...
Merge infershapecontext and ExecutionContext
7 years ago
c464ec21d8
Fix bug of foward default attribute not passed to backward
7 years ago
0ff540cc7f
Merge pull request #4640 from Canpio/fix_flush_var
...
Fix flushing vars
7 years ago
4cb5bd9021
Implementing the Adamax optimizer operator ( #4538 )
...
* Implementing the Adamax optimizer step operator
* Adding unit tests for adamax_op
* Changing learning rate and time step to inputs from attributes
* Changing learning rate and time step to input(tensors)
* Making the Adamax operator conform to naming convention
* Removing Tensor<float> from comments
* Rectifying the Adamax implementation
* Changing Unit Test values and adding comments
* Changing Unit Test to test multiple steps
7 years ago
f30a1f42f0
Adding relu6 activation function ( #4607 )
7 years ago
b14c122a94
Fix bug
7 years ago
d9585f9aa4
Merge pull request #4632 from luotao1/reduce
...
Unify Reduce functions and simplify register code
7 years ago
597299074e
fix bug in REGISTER_OP(reduce_min)
7 years ago
a06f099d9f
refine comment of interp_op
7 years ago
fcfce48421
follow coments
7 years ago
63309941b3
pull develop and update
7 years ago
282435204e
Merge branch 'develop' of https://github.com/PaddlePaddle/Paddle into auc_op
7 years ago
53574e54a1
reslove merge conflict;reimplement ELU activation with functor
7 years ago
707d144c93
Unify Reduce functions and simplify register code
7 years ago
aa1e92ec9a
Merge branch 'develop' of https://github.com/baidu/Paddle into inference
7 years ago
8f2c48cf4c
Fix Layer.cpp
7 years ago
5b862fedf1
remove debug log in interp_op.cc
7 years ago
089cc11df4
clean up && fix #4624
7 years ago
adad8d9ed2
Open WITH_TESTING option.
7 years ago
4724bdbe68
Merge branch 'develop' into interp
7 years ago
3db3a1066b
remove conflict
7 years ago
ba791f7b3f
Add vol2col functor and unit test
7 years ago
dcfbbd3f1d
Merge branch 'develop' into crf
7 years ago
c93d74aa06
merge develop
7 years ago
6e7666f199
before backward
7 years ago
c83ea1cdca
remove hardcode add_XX_op
7 years ago
a67e8ea3eb
Add AddOp
7 years ago
005f15b495
FeedOp and FetchOp unit test
7 years ago
c0a34e1c64
rename InferShapeContextBase to InferShapeContext
7 years ago
a0767228bd
merge InferShapeContext and ExecutionContext
7 years ago
c3b46d1683
Merge pull request #4573 from Canpio/dev_backward_for_op_desc
...
Backward for op desc
7 years ago
cb1baa3be7
Update program.md ( #4618 )
...
* Update program.md
* Update
* Update
7 years ago
99895730f7
Merge pull request #4609 from kavyasrinet/tanhshrink
...
Implementing tanhshrink operator
7 years ago
b68a95f7f4
prune pass simple test
7 years ago
20a6ae7f1f
Feature/tensor array add python binding ( #4616 )
7 years ago
683ef60d7c
Merge commit 'refs/pull/4537/head' of https://github.com/PaddlePaddle/Paddle into 4537
7 years ago
a7d700e0ba
revert local scope to TODO
7 years ago
bd7b66920c
Merge branch 'develop' of https://github.com/PaddlePaddle/Paddle into dev_backward_for_op_desc
7 years ago
4c96008a45
Merge pull request #4566 from reyoung/feature/grad_reg_mechanism_cont2
...
Complete register gradient for compile time
7 years ago
f087533cc3
Merge remote-tracking branch 'baidu/develop' into executor_impl
7 years ago
91f5d2b9cb
follow comments and create local_scope inside executor run method
7 years ago
803b7b6298
Follow comments
7 years ago
f8b5d54c6c
Merge pull request #4569 from jacquesqiao/add_compile_time_infershape
...
Add compile time infershape
7 years ago
d73aa874b2
merge develop
7 years ago
097f533bca
Resolve conflict
7 years ago
e8a678e1ee
fix executor gpu unittest runtime error
7 years ago
ac0e3828b4
test text
7 years ago
4acd5abaaa
update comment for input/output length check
7 years ago
3e82922851
fix tensorarray unpack bug ( #4614 )
7 years ago
8c7ee7a5cd
optimize code
7 years ago
1f5192a27b
fix executor gpu unittest
7 years ago
ffe1b69229
Merge branch 'develop' of https://github.com/PaddlePaddle/Paddle into add_compile_time_infershape
7 years ago
fe6962e0b5
Merge pull request #4600 from reyoung/feature/add_persistable_in_var_desc
...
Add persistable in framework.proto
7 years ago
087addaa76
Merge pull request #4558 from kexinzhao/adagrad_op
...
Implementing the Adagrad optimizer step operator
7 years ago
78f4c803f3
change learning rate and fix format
7 years ago
39f75a13a4
Merge remote-tracking branch 'baidu/develop' into executor_impl
7 years ago
bbceb72398
refine some codes
7 years ago
48b080db9f
ensure global BuddyAllocator is initialized before global Scope
7 years ago
f52cdaa0ce
Updated RMSProp to have learning rate as an input and work with GPU
7 years ago
0336304176
Merge branch 'develop' of https://github.com/PaddlePaddle/Paddle into rmsprop
7 years ago
154a6ed29c
Implementing tanhshrink operator
7 years ago
db77937ea4
Fix learning_rate usage for momentum
7 years ago
45c4dcaabb
add fetch operator
7 years ago
e043b38660
clean code
7 years ago
628715d602
clean code
7 years ago
352af966d7
add python unit test
7 years ago
3e2be065b9
Merge pull request #4604 from kavyasrinet/activations
...
Added Leaky Relu activation
7 years ago
20725f2d52
add executor feed operator test
7 years ago
c10da26cf5
Modify implementation
7 years ago
4b07686aa8
Add unit tests
7 years ago
828c5b3e1d
Adding Adadelta optimization operator ( #4576 )
...
* Adding Adadelta optimization operator
* Making inputs and outputs conform to naming convention
* Removing type alias from header files
* Fixing Adadelta documentation in comments
* Addressing code review feedback
7 years ago
60af56c1b8
Added Leaky Relu activation
7 years ago
623848afa1
add feed operator
7 years ago
a598ef5388
Merge branch 'feature/grad_reg_mechanism_cont2' of https://github.com/reyoung/Paddle into dev_backward_for_op_desc_dev
7 years ago
1172f24929
Merge pull request #4590 from wangkuiyi/paddle_only_cpu
...
Add -D PADDLE_WITH_CUDA in cmake/configure.cmake
7 years ago
c0511c8a93
Merge pull request #4598 from jacquesqiao/fix-sgd-learning-rate
...
use EigenVector to get learning_rate for GPU device in SGD operator
7 years ago
3014f6a113
correct macro
7 years ago
8ebc31d935
optimize the dsize
7 years ago
775c60246b
remove using in sgd header file
7 years ago
cc1860c10e
Add persistable in framework.proto
7 years ago
ebbbaee0c3
Follow comments
7 years ago
2594a50245
Polish code
7 years ago
a270dbb778
Add support for rnn_op
7 years ago
c4effc7d2d
Fix CI Test
7 years ago
ee7b3ed09e
use EigenScalar to get learning_rate from GPU device
7 years ago
880b874b47
Merge branch 'develop' of https://github.com/paddlepaddle/paddle into paddle_only_cpu
7 years ago
5917e09cde
tmp work
7 years ago
2b204f048b
Rename platform::GetDeviceCount into platform::GetCUDADeviceCount
7 years ago
e02cc571cf
Merge remote-tracking branch 'baidu/develop' into executor_impl
7 years ago
fe10e86dd5
fix gpu build error
7 years ago
473ca534b0
Merge pull request #4579 from wangkuiyi/update_lod_tensor_design
...
Update lod_tensor.md
7 years ago
4558807c48
Use PADDLE_WITH_CUDA instead of PADDLE_WITH_GPU
7 years ago
e79d2f1b65
Merge pull request #4584 from reyoung/feature/change_macro_paddle_no_gpu
...
Change `PADDLE_ONLY_CPU` to `PADDLE_WITH_GPU`
7 years ago
fa12e51675
Adding the default attribute test case
7 years ago
94855f4af0
Fixed changes proposed in the review
7 years ago
e119177a8c
Use unique_ptr
7 years ago
c61e82bc21
Merge branch 'develop' of https://github.com/PaddlePaddle/Paddle into dev_backward_for_op_desc
7 years ago
84500f9487
Change `PADDLE_ONLY_CPU` to `PADDLE_WITH_GPU`
...
By shell command
```bash
sed -i 's#ifdef PADDLE_ONLY_CPU#ifndef PADDLE_WITH_GPU#g' `find ./paddle/ -name '*.h' -o -name '*.cc' -o -name '*.cpp' -o -name '*.c' -o -name '*.cu'`
sed -i 's#ifndef PADDLE_ONLY_CPU#ifdef PADDLE_WITH_GPU#g' `find ./paddle/ -name '*.h' -o -name '*.cc' -o -name '*.cpp' -o -name '*.c' -o -name '*.cu'`
```
7 years ago
37b0bb1597
Fix compile errors
7 years ago
eed2c1e1d6
Changing SGD inputs and outputs to conform to Operator naming convention ( #4586 )
7 years ago
77e07833f5
Merge pull request #4582 from Canpio/fix_bugwq
...
Bug fix
7 years ago
ab9545aa95
add shape_inference_map
7 years ago
Yu Yang
tensor-tang
zchen0211
qijun
Abhinav Arora
hedaoyuan
dangqingqing
武毅
Yibing Liu
wanghaoshuang
Yang Yang(Tony)
Dong Zhihong
kavyasrinet
fengjiayi
chengduoZH
Markus Kliegl
dzhwinter
Qiao Longfei
Luo Tao
caoying03
chengxingyi
Guo Sheng
Yan Chunwei
guosheng
kexinzhao
Helin Wang
Cao Ying
Yancey1989
chengduo
Liu Yiqun
Siddharth Goyal
zhouxiao-coder
Abhinav Arora
gangliao
Yi Wang
Yi Wang
Kexin Zhao