9eec2c7509
refine pe
7 years ago
170ac721b6
remove unnecessary tensor copy in save op
7 years ago
9ca8124f99
Merge branch 'develop' of https://github.com/PaddlePaddle/Paddle into develop
...
yes
7 years ago
8b1918f5c7
Merge pull request #10454 from chengduoZH/fix_fetchop
...
Fix fetch_op_handle
7 years ago
2bff03bc1e
fix a compile error ( #10488 )
7 years ago
0ae97e8a5b
Merge branch 'develop' into refine_relu_test
7 years ago
f4851f14b8
clean code
7 years ago
a459764de7
Merge branch 'develop' of https://github.com/PaddlePaddle/Paddle into fix_fetchop
7 years ago
e00c1ee10f
fix split var test
7 years ago
0e78cb69fb
Clean OpProtoAndCheckerMaker
...
Do not use ctor
* Reduce line of codes.
* We can use virtual function for Maker now.
* The implementation does not care what maker holds, it is easier to
refactor later.
7 years ago
89dcb0bd15
refine EngineIOConverter, and use io_convert in test_trt_activation_op
7 years ago
96b703c007
Merge branch 'develop' of https://github.com/PaddlePaddle/Paddle into feature/clean_matmul
7 years ago
8cbb49ce63
rm space
7 years ago
e97645c959
rm block_desc.h whitespace
7 years ago
345737d0fe
add sync
7 years ago
fcd31d6161
Follow comments and polish code names
7 years ago
187e23a79c
fix MatMul parameter
7 years ago
0a13d3c67a
Move MatMul to blas_impl.h
...
Rename MatDim to MatDescriptor
7 years ago
be1f058f52
Merge branch 'develop' into develop
7 years ago
55218dafbc
add prependAllocatedop
7 years ago
3356fb3c6e
Merge pull request #10461 from luotao1/refine_convert
...
refine io_convert and op_convert
7 years ago
676dfd1875
follow comments
7 years ago
a135fec1fc
Merge branch 'develop' of https://github.com/PaddlePaddle/Paddle into gen_nccl_id_op
7 years ago
aff8a26d71
check generated_op_
7 years ago
0f86397d81
fix build
7 years ago
53b401d589
refine io_convert and op_convert
7 years ago
2e5d44f102
fix fetch op
7 years ago
3dd01823a8
Merge branch 'develop' of https://github.com/PaddlePaddle/Paddle into feature/clean_matmul
7 years ago
17009d0627
workable version
7 years ago
dce0732d5e
Merge pull request #10380 from panyx0718/dist_timeline
...
timeline for distributed training
7 years ago
0c518888fa
Merge pull request #10430 from panyx0718/infer
...
Add comment to explain how to run inference test
7 years ago
c6a6d87f96
Rewrite Matmul, make code cleaner
7 years ago
a529d790b6
Merge branch 'develop' of https://github.com/PaddlePaddle/Paddle into gen_nccl_id_op
7 years ago
2a2c83b9e6
feature/convert tensorrt io ( #10440 )
...
* init
* init
* add ut
* split singleton from base class
* add singleton
* ad singleton
7 years ago
82c61dbde3
fix testing
7 years ago
9fccf46270
reword comments
7 years ago
d1ea74d3b9
follow comments
7 years ago
b708ec0ae1
Merge pull request #10412 from JiayiFeng/correct_TensorCopy_misuse
...
Correct tensor copy misuse
7 years ago
55e714e0d2
add float16 support to pool3d
7 years ago
8b16927230
add fp16 support to conv3d
7 years ago
fd1971caa0
Add the call of DropKids at the end of executor.Run to delete the local scopes created in operators ( #10403 )
...
* Add the call of DeleteScope to delete the memory of scope created by NewScope.
* Call DropKids at the end of executor.Run to delete all local scopes created in operators.
7 years ago
99acf1da4c
Merge pull request #10351 from chengduoZH/feature/update_sparse_parameter
...
Feature/update sparse parameter
7 years ago
8f8a4768dc
adding device_context to blas deps list ( #10420 )
...
* adding operator to blas deps list
* use device_context instead to solve cycle deps
7 years ago
a28dffbb0b
Fix/adam float64 ( #10407 )
...
* "optimizer op support float64"
* "fix ci"
* "fix ftrl op"
7 years ago
cdd52f3a30
Add comment to explain how to run inference test
7 years ago
0598a4b366
fix ci
7 years ago
3667578ec2
testing
7 years ago
881e063ee2
follow comments
7 years ago
ff599b9218
use Reduce and Broadcast
7 years ago
0441c2cc45
fix ci
7 years ago
0c99cd7bbb
fix errors in sequence_padding_test
7 years ago
6418c42148
Travis: fix check style error.
7 years ago
4e3fac4129
fix sign unsigned comparison ( #10424 )
7 years ago
b65282168c
Fix cpplint errors in lstm kernel ( #10394 )
7 years ago
4558c0ec0a
Merge pull request #10414 from chengduoZH/wrap_shfl_x_sync
...
Wrap shfl_x_sync
7 years ago
d36af62c1e
wrap_shfl_x_sync
7 years ago
2d98a418d7
fix remove op ( #10410 )
...
* fix remove op
* update
7 years ago
bf99396a04
fix errors in sequence_slice_op
7 years ago
baa9f50da5
fix errors in multiplex_op
7 years ago
2e617334eb
fix errors in lod_reset_op
7 years ago
e309f42293
fix errors in concat_test
7 years ago
0285a2b95d
Merge pull request #10371 from reyoung/refine_code
...
Polish MatMul, clean copy & paste code
7 years ago
f9c680c43e
Merge branch 'develop' of https://github.com/PaddlePaddle/Paddle into feature/update_sparse_parameter
7 years ago
d9320dcd94
complete code
7 years ago
7722baa8e3
follow comments and clean code
7 years ago
5a9f17f02b
clean up
7 years ago
4646c0f35d
Merge pull request #10144 from luotao1/tr_convert_init
...
tensorrt convert init
7 years ago
3bb99c4f66
Added auto transform to beam_search_decode_op ( #10286 )
...
* Added auto transform to beam_search_decode_op
* Added some comment
* Added unittest for beam_search_decode_op on GPU
7 years ago
c9f55dfafc
Fix CPPLint issues in /math/detail/gru_kernel.h ( #10390 )
...
* Fix CPPLint issyes in gru_kernel.h
* Fix CPPLint issyes in gru_kernel.h
* Fix Compile error
7 years ago
ef6ea790dc
Clean and extract blas
7 years ago
ccc594e4c4
need to copy LoD info ( #10392 )
7 years ago
7a86069422
Add float16 demo code and put float16 work in contrib/float16 folder ( #10331 )
...
* add test float16 inference accuracy example
* complete the test
* clean code
* add argument parse and refine tests
* add shell script
* add float16 benchmark code
* refine code
* prepare for contrib/float16
* put things in contrib float16 folder
* update benchmark result
* further update benchmark report
* add float16 inference report
* update report
7 years ago
beb1245560
add relu converter and unit-test
7 years ago
76d8b14bce
Add timeline support for distributed training
7 years ago
d0785ce982
Merge branch 'develop' of https://github.com/PaddlePaddle/Paddle into refine_code
7 years ago
815d888468
Clean MatMul
7 years ago
2abcf37958
Merge pull request #10327 from reyoung/feature/clean_blas
...
Feature/clean blas
7 years ago
54797abd53
Merge pull request #10347 from chengduoZH/replace___shfl_with__shfl_sync
...
Wrap __shfl
7 years ago
62fed4cbb3
fix __shfl_down ( #10362 )
7 years ago
9d7279b953
Merge branch 'develop' of https://github.com/PaddlePaddle/Paddle into refine_code
7 years ago
bc8160350b
Fix compile
7 years ago
e97c1a8ca0
fix __shfl
7 years ago
a6edeb39b3
Merge branch 'develop' of https://github.com/PaddlePaddle/Paddle into feature/clean_blas
7 years ago
7237323c5d
fix compile
7 years ago
f63ff90b03
Fix/fp64 ( #10346 )
...
* "fix double type error"
* "fix ci"
* "softmax fp64"
* "fix momentum"
* "fix ci"
7 years ago
88d79dfe95
Merge pull request #10292 from typhoonzero/fix_grpc_server_ready_condition
...
Fix grpc server ready condition
7 years ago
0cc635497c
merge develop
7 years ago
6084af47ef
Fix the bug when a input variable of op is dispensable. ( #10268 )
...
* Fix the bug when a input variable of op is dispensable.
* Add HasInputs/Outputs interfaces to OperatorBase.
* Remove the unreferenced header file.
7 years ago
8a0c7e2e70
Merge pull request #10280 from reyoung/feature/add_stable_test_of_cross_entropy
...
Clean cross entropy
7 years ago
4a497b826d
MKLDNN implementation of batch normalization ( #9904 )
...
* Initial implementation of forward pass for MKLDNN batch norm
* Added attributes for MKLDNN batch norm
* MKLDNN batch norm forward pass passes unittest. Started working on backward
* Backward pass for MKLDNN batch norm added
* MKLDNN batch norm: scoring added to forward pass
* MKLDNN batch norm: bias as input added; handling AnyLayout when kernel is looked up
* MKLDNN batch norm: python unit tests added; mkldnn tests removed
* MKLDNN batch norm: changes required by cpplint
* MKLDNN batch norm: refactoring the operator
* MKLDNN batch norm: saved variance inversed in backward pass for correct execution of MKLDNN unit tests
* MKLDNN batch norm: refctoring, function for static/const cast to void* added
* MKLDNN batch norm: remove AnyLayout from batch norm
* MKLDNN batch norm: only NCHW format is supported. Unittests refactored
* MKDNN batch norm: use_mkldnn added to attributes
* MKLDNN batch norm: AnyLayout removed from unittest
* MKLDNN batch norm: added CUDNN defines to batch norm
* MKLDNN batch norm: undefined data_format variable corrected
* MKLDNN batch norm: use_cudnn added, use of setUp method for configuring attributes
* MKLDNN batch norm: added use_cudnn attribute to batch norm operator
* MKLDNN batch norm: correcting batch norm unit tests for MKLDNN
* MKLDNN batch norm: MKLDNN tests moved to another file; reverting changes for saved variance not being inverted
* Change default layout to NCHW
* MKLDNN batch norm: init_kernel_type method added to unit tests
* MKLDNN batch norm: style changes
* MKLDNN batch norm: unit tests refactored
* MKLDNN batch norm: added use_mkldnn attribute to batch norm python interface
7 years ago
4fbde42cdf
Fix __shfl_down_sync_ of cross_entropy ( #10345 )
...
* fix __shfl_down_sync_ of cross_entropy
* use reduceSum
* "fix ci"
7 years ago
c891189568
update sparse gradient parameter with reduce and broadcast
7 years ago
6422c0e4f6
update by comment
7 years ago
b8f7fa97b6
replace __shfl with __shfl_sync
7 years ago
5ff1ef36ee
update sparse parameter
7 years ago
eeed7af5c3
add gen_nccl_id_op
7 years ago
5e151b2c83
Follow comment
7 years ago
9a4c1a39f0
Merge branch 'develop' of https://github.com/PaddlePaddle/Paddle into feature/add_stable_test_of_cross_entropy
7 years ago
caa4027d9d
Follow comments
7 years ago
57be5c6c74
"fix double type error" ( #10322 )
...
* "fix double type error"
* "fix ci"
7 years ago
faebadd938
Merge pull request #10228 from jacquesqiao/use-multi-thread-todo-update
...
Use multi thread to do update
7 years ago
ff99d94197
Merge pull request #10164 from Yancey1989/lookup_sparse_table_op
...
add lookup_sparse_table_op
7 years ago
b3cf429e02
Merge branch 'develop' of https://github.com/PaddlePaddle/Paddle into fix_grpc_server_ready_condition
7 years ago
e7ac709b4b
done
7 years ago
1945b729b6
Fix CPPLint issues with math/sequence_padding ( #10317 )
...
* Fix cpplint issues in sequence_padding
* Fix typo in cu file
* Fix dependencies of sequence_padding
* Add include
7 years ago
9bcd9f661b
fix cpplint error ( #10329 )
7 years ago
55f0d84029
Fix Cpplint Issues in fluid/inference/tensorrt/ ( #10318 )
...
* Fix CPPLint issues in fluid/inference/tensorrt/
* Fix compile errors
7 years ago
4db43c6c9f
Naive implement cblas
7 years ago
a1a401eb26
fix
7 years ago
d11b8e56e5
fix
7 years ago
9945265f09
Merge branch 'develop' into tr_convert_init
7 years ago
a131c73fcf
Merge branch 'develop' of https://github.com/PaddlePaddle/Paddle into fix_grpc_server_ready_condition
7 years ago
3222cf16f7
Merge pull request #10325 from chengduoZH/fix_shfl_sync
...
Fix shfl_sync for CUDA8.0
7 years ago
4613aeba0e
Merge pull request #10272 from kexinzhao/save_fp16
...
Add float16 support to save op
7 years ago
60d6348e69
Revert develop
7 years ago
86af6bdc81
Merge branch 'develop' of https://github.com/PaddlePaddle/Paddle into feature/clean_blas
7 years ago
9a8be9daca
Merge pull request #10223 from pkuyym/fix-10219
...
Change `customize_loss_grad` to `use_default_grad_scale`.
7 years ago
49dedfad17
Polish code and tests
7 years ago
90d73c79c3
fix shfl_sync for CUDA8.0
7 years ago
d86626df84
optimize log
7 years ago
ebf0027391
use IOThreadPool to dispatch async update task
7 years ago
ea372b3452
add more log
7 years ago
738585476d
Fix more CPPLint issues in fluid/operators/math ( #10276 )
...
* Fix CPPLint issues in lstm_cpu_kernel.h
* Fix CPPLint issues in math/math_function_test
* Fix CPPLint issues in math/math_function_test
* Fix CPPLint issues in math/concat.cc
* Fix CPPLint issues in math/concat.cc
* Fix CPPLint issues in math/concat.cc
* Fix CPPLint issues in math/gru_cpu_kernel
* Fix CPPLint issues in math/selected_rows_functor_test.cu
* Fix compile error
* Fix compile error
7 years ago
d25fdb0a47
fix build: cuda_helper.h not found
7 years ago
eb6f9dd5de
Feature/cuda9 cudnn7 ( #10140 )
...
* "re-commit "
* "picked up"
* "fix ci"
* "fix pdb hang up issue in cuda 9"
7 years ago
008f6df9b2
update
7 years ago
f61dfeedcc
Merge pull request #10263 from chengduoZH/add_FLAGS_use_deterministic_algo
...
Add FLAGS_cudnn_algo_use_autotune
7 years ago
ef48f3c766
wip
7 years ago
c888e01660
Refactor GEMM in blas
7 years ago
13fac4232a
Fix to pass CI.
7 years ago
c0ac0cd6b3
Complete rename
7 years ago
79be1bb3df
Merge branch 'develop' into fix-10026
7 years ago
48466b4424
auto => auto*
7 years ago
6c18410487
Revert code to develop
7 years ago
9fda5c92cd
Merge branch 'develop' of https://github.com/PaddlePaddle/Paddle into add_FLAGS_use_deterministic_algo
7 years ago
f456cd8079
Merge branch 'develop' of https://github.com/PaddlePaddle/Paddle into fix-10219
7 years ago
76174ec0e9
Clean cross entropy and add sync in executor
7 years ago
25779c982d
Merge pull request #10265 from reyoung/feature/polish_code
...
Add comments and polish code
7 years ago
c5774e3282
add FLAGS_use_deterministic_algo
7 years ago
e735359631
Fix more CPPlint issues in fluid/operators/math ( #10249 )
...
* Fix CPPLint errors
* Fix CPPLint errors in sequence2batch
* Fix compilation
* Fix LSTM op and GRU op
* Fix LSTMP op
* Fix more cpplint errors in operators/math
* Address Code review feedback
7 years ago
efba1c7dcb
address comments
7 years ago
6c88f1ae6e
add save op float16 support
7 years ago
f82cb635cf
optimize code, add more log
7 years ago
9612c7e599
Add comments and polish code
7 years ago
deabc8ca0b
Merge branch 'develop' into feature/clean_memcpy_async
7 years ago
71fa3ca9c4
Merge pull request #10232 from JiayiFeng/fix_unittests
...
Fix unittests
7 years ago
76c4ae856f
Fix reshape op. ( #10253 )
7 years ago
1a93253f16
fix unittest
7 years ago
30f9dc92e5
fix errors
7 years ago
2f9fa9b721
Merge pull request #10167 from wanghaoshuang/fluid_init
...
Add init interface for customize devices.
7 years ago
6f6f330423
update the register method
7 years ago
9997c916fc
Pull origin
...
Merge branch 'fix-10026' of github.com:ktlichkid/Paddle into fix-10026
7 years ago
709a9edd46
Code clean up
7 years ago
330fa95cbd
Follow comments
7 years ago
4a5bfa89c3
Modify RoI pooling op to use LoDTensor and expose it into Python API ( #10208 )
...
* modify roi pool with lod and expose ROI Pooling into Python API
* make lod code brief
* make doc more clearly
* make doc more clearly
7 years ago
e498e1fc56
Adam operator optimized with Eigen ( #10229 )
...
* Some changes for Adam profiling
* Adam optimization: initial Eigen optimization
* Eigen Adam: flavour of adam can be chosen
* Eigen Adam used for CPU by default. Plain Adam used for GPU
* Eigen Adam: missing call to the Eigen functor added
* Eigen Adam: revert changes in benchmarks
* Eigen Adam: typo corrected
7 years ago
0ecc6fa8f3
Add float16 transpiler and image classification example ( #10109 )
...
* add float16 transpiler
* fix feed fetch target names mismatch
* fix cast op input change issue
* fix program desc flush error
* fix inconsistent var names in block desc bug
* code clean up
* add float16 infernce C++ example and fix prune bug
7 years ago
83b1a8f6bf
Pending more CPPLint errors in fluid/operators/math ( #10243 )
...
* Fix CPPLint issue in test_engine
* Fix CPPLint errors in operators/math
* Fix compilation
7 years ago
c816121d11
optimized iou_similarity_op ( #10231 )
7 years ago
b88721213f
fix broadcast_op_test and reduce_op_test
7 years ago
bcf260e1e8
fix several unit tests
7 years ago
b058189941
Merge branch 'develop' of https://github.com/PaddlePaddle/Paddle into use-multi-thread-todo-update
7 years ago
6d934560c7
Merge pull request #10042 from jacquesqiao/add-async-listen-and-serv-op
...
listen_and_serv_op support async update
7 years ago
f457d5da06
Fix more CPPLint errors ( #10218 )
...
* Fix more CPPLint issues
* Fix more CPPLint issues
* Fix more CPPLint issues
* Fix CPPLint issues in operators/math and operators/reader
7 years ago
0d491b670a
use-multi-thread-todo-update
7 years ago
3295f31076
optimize naming
7 years ago
18d6254d44
Merge branch 'develop' of https://github.com/PaddlePaddle/Paddle into fix-10219
7 years ago
46342a2306
delete useless code
7 years ago
848fb00215
Fix comments.
7 years ago
9c7fa6ff69
Merge pull request #10206 from JiayiFeng/blocking_queue_for_reader
...
Blocking queue for reader
7 years ago
0264ec3957
Merge branch 'develop' of https://github.com/PaddlePaddle/Paddle into add-async-listen-and-serv-op
7 years ago
63bd38bd74
code optimize
7 years ago
c02ba51de0
Merge pull request #10191 from reyoung/feature/strict_dynload
...
Make dyload strictly use the same ABI in header
7 years ago
dccd013bd3
refine distribute transpiler
7 years ago
8bd34664f1
fix unit test error
7 years ago
17c51d69d1
fix unit test error
7 years ago
304b6b7138
Follow comments
7 years ago
82571deb89
Change `customize_loss_grad` to `use_default_grad_scale`.
7 years ago
326221acec
Merge branch 'develop' into tr_convert_init
7 years ago
4c8ff72615
Fix CPPLint errors with rxecutor ( #10212 )
7 years ago
5fe1fe3a27
Fix signed/unsigned comparison warning ( #10211 )
7 years ago
4cb63d8451
Remove unnecessary header files
7 years ago
e393c86c4a
Merge branch 'develop' of github.com:PaddlePaddle/Paddle into lookup_sparse_table_op
7 years ago
c4e3010b14
use template to do registry
7 years ago
2d57158e2b
fea/init tensorrt engine ( #10003 )
7 years ago
d599de5c41
auto registray op converters
7 years ago
e057ba6877
Merge branch 'develop' of https://github.com/PaddlePaddle/Paddle into blocking_queue_for_reader
7 years ago
e2ca42408b
Replace Channel in DoubleBufferReader with BlockingQueue
7 years ago
a786611314
Replace Channel in MultiFileReader with BlockingQueue
7 years ago
1a25f3cd07
Add reader blocking queue
7 years ago
0c24b3f937
Clean memcpy async
7 years ago
64babc9aeb
Merge pull request #10189 from reyoung/feature/fix_matmul_bug
...
Fix batch_gemm bugs
7 years ago
580dad0c2c
Fix compile when there is no mkl
7 years ago
3d53631bad
Make dyload strictly use the same ABI in header
7 years ago
2a06e307d0
Fix batch_gemm bugs
...
stride should be int64_t, not int
7 years ago
8aea5cac0a
add attr auto_grown_table
7 years ago
2f53cd0a76
Fix beam_search memory leak. ( #10185 )
7 years ago
5afc2a9972
Keep up with upstream
...
Merge remote-tracking branch 'upstream/develop' into fix-10026
7 years ago
63bf82ddea
Merge branch 'develop' into add-async-listen-and-serv-op
7 years ago
8b2391858f
Merge pull request #10181 from abhinavarora/cpplint_advanced
...
Fix CPPLint errors with framework/op_desc
7 years ago
3fdfa940be
Merge pull request #10135 from typhoonzero/unify_blocking_queue
...
Unify fluid blocking queue
7 years ago
ad3f6f4ad5
Fix devices 'not undefined' error.
7 years ago
edd3587e50
Fix CPPLint errors with op_desc
7 years ago
81dfc0cf0e
Clean up unused code in operator class ( #10035 )
...
* delete unused IsNetOp() and Rename()
* rm OperatorBase::Rename implementation
* delete Operator::InputVars()
* remove unused OperatorBase::ShareLoD; ShareLoD has been implemented in infershape
* organize operatorbase; remove unused set_type
* add comments
* fix comment
7 years ago
f09aed0475
Fix CPPLint issues in framework/data_transform framework/prune.cc ( #10178 )
...
* Fic CPPLint issues with data_transform
* Fic CPPLint issues with prune.cc
7 years ago
72ee737f3f
Merge pull request #9308 from wangyang59/bilinear
...
Bilinear interp op
7 years ago
2182ecfbbd
remove duplicated ShareLoD in gru_op and sequence_conv_op ( #10149 )
...
* remove share lod; it has already been performed in infershape
* slightly release test bound of test_image_classification
7 years ago
6171705a2c
Potential bug in paddle/fluid/platform/CMakeLists.txt ( #9723 )
...
* fix
* nv_library
* add with_gpu
* revert
7 years ago
fc025f5265
Fix memory leak of pserver ( #10173 )
7 years ago
3d96b3811a
Fix InitGflags.
7 years ago
48473dddf4
Merge branch 'develop' into tr_convert_init
7 years ago
a4b452a2d6
Remove initP2P(bool) and init function in framework.
7 years ago
4ecc9b7bae
Merge pull request #10166 from reyoung/feature/train_and_test_recordio
...
Add demo for recordio train/test and parallel executor
7 years ago
e4708565f4
Fix cpplint format.
7 years ago
a0b258278e
Reuse 'initP2P(bool, std::vector)' in 'initP2P(bool)'
7 years ago
f31bb1476c
Merge branch 'develop' of https://github.com/PaddlePaddle/Paddle into fluid_init
7 years ago
5a433ecb60
Merge pull request #10134 from luotao1/tensorrt_include
...
refine tensorrt cmake and dockerfile
7 years ago
bfbbe19fbb
Merge pull request #10150 from chengduoZH/fix_elementwise_gradient
...
Fix elementwise_gradient bug
7 years ago
48b7b54321
Refine code.
7 years ago
5ce57555ee
Fix CPPLint issues in init.cc, init.h and library_type.h ( #10148 )
...
* Fix CPPLint issues in init
* Fix compilation
* Fix typo in init.cc
* Fix CPPLint issues in library_type.h
* Fix compilation in init.h
7 years ago
0f5d5b1ffc
Merge branch 'develop' of https://github.com/PaddlePaddle/Paddle into fix_elementwise_gradient
7 years ago
1bdea0a8d2
Add init interface for customize devices.
7 years ago
8081e15774
fix send_recv_op_test
7 years ago
2c8fe4e544
Merge pull request #10143 from typhoonzero/fix_multiGPU_dist_train
...
Fix multi gpu dist train
7 years ago
e5f2cb8021
Merge pull request #10160 from JiayiFeng/develop
...
fix fetch_op_handle
7 years ago
54ada9449e
Add demo for recordio train/test and parallel executor
7 years ago
44fa823841
Merge pull request #9949 from mozga-intel/mozga-intel/Mul_mkldnn
...
Initial implementation of multiplication operator for MKLDNN
7 years ago
e8d802159e
add lookup_sparse_table_op
7 years ago
d06c79c7a7
fix elementwise_grad op kernel and add unit test
7 years ago
00b54d5803
fix fetch_op_handle
7 years ago
dd5946528a
Merge pull request #10142 from JiayiFeng/Add_TensorCopySync
...
Add synchronous TensorCopy
7 years ago
2486d563ba
Create README.md of fluid/recordio ( #10145 )
...
* Create README.md
* Update README.md
7 years ago
469a349ae3
polishing after qingqing's comments
7 years ago
7436b36875
make bilinear_op registration up-to-date
7 years ago
4a3c99f334
after rebase
7 years ago
d61738311a
remove dropout and nccl test due to frequent CI failures
7 years ago
3e6718e2de
simplified include structure
7 years ago
d87ac4de34
GPU of bilinear_interp_op done
7 years ago
ad3b3d9dc1
ported old paddle gpu bilinear_interp
7 years ago
67ce586453
gpu implementation of bilinear interp
7 years ago
f67f0cae50
finished testing cpu bilinear_interp_op
7 years ago
c7cd6d130b
cpu implement of bilinear interp
7 years ago
504e60a881
Fix Cpplint issues in framework/data_type.h and framework/feed_fetch_type.h ( #10146 )
...
* Fix CPPLint issues with data_type.h
* Fix CPPLint issues with feed_fetch_type.h
7 years ago
3503c47f9a
listen and serv default sync mode
7 years ago
0881d809cd
Merge branch 'develop' of https://github.com/PaddlePaddle/Paddle into add-async-listen-and-serv-op
7 years ago
42febfa928
tensorrt convert init
7 years ago
f03415207a
Merge branch 'develop' of https://github.com/PaddlePaddle/Paddle into fix_multiGPU_dist_train
7 years ago
1fba0c578a
fix multi gpu dist train
7 years ago
c5e178f4fd
fix errors
7 years ago
9f11da5931
Add synchronous TensorCopy and use it in double buffer
7 years ago
64509fd93b
Style fix
7 years ago
294b58a9ba
Changed registered type
7 years ago
df80b6ea8c
Added InferVarType
7 years ago
f57efeb6d1
Added GetExpectedKernelType and Debug message
7 years ago
6f06b32258
Added GetExpectedKernelType and Debug message
7 years ago
f3e4e42def
Merge pull request #10130 from reyoung/feature/skip_loss
...
Add customize_loss_grad option to PE
7 years ago
a29e352b80
optimize code
7 years ago
a0ced3df82
async update can run
7 years ago
55feba9b5a
Fix CPU compile
7 years ago
3863c6a9a8
Merge pull request #10126 from typhoonzero/fix_para_exe_dist_train
...
Fix para exe dist train
7 years ago
251e4a8ee5
unify fluid blocking queue
7 years ago
7a395881d4
Add customize_loss_grad option to PE
7 years ago
71f51ff64a
refine tensorrt cmake and dockerfile
7 years ago
42a15a43b7
Merge branch 'develop' of https://github.com/PaddlePaddle/Paddle into add-async-listen-and-serv-op
7 years ago
2b06b4b4e9
updates follow up para exe
7 years ago
8ee837255e
fix send op handle local scope
7 years ago
63055a3e08
complete grad_to_id
7 years ago
6c0356e41d
Merge pull request #10119 from chengduoZH/rollback_delay_op
...
Enable delay op feature
7 years ago
f5b4ac6e50
enable delay op feature
7 years ago
8023c6d749
Create sub socpe when it is necessary
7 years ago
4b86b49ecd
Merge branch 'fix-build-activation_op' of ssh://github.com/jacquesqiao/Paddle into add-async-listen-and-serv-op
7 years ago
108e71cc94
fix build activation_op.cc on mac
7 years ago
c6937abdd1
tmp
7 years ago
58fd4c0994
Merge pull request #10111 from abhinavarora/cpplint_tensor_util_test
...
Fix CPPLint issues in tensor_util_test
7 years ago
1d75674614
Merge branch 'develop' of https://github.com/PaddlePaddle/Paddle into add-async-listen-and-serv-op
7 years ago
1c70600d09
Fix compile error
7 years ago
6402b59a7c
Fix CPPLint issues in some tests in fluid/framework ( #10068 )
...
* Fix CPPLint in data_device_transform_test
* Fix compilation error
* Fix compilation error
* Fix CPPLint errors in data_layout_transform_test
* Fix CPPLint errors in data_type_transform_test
* Fix CPPLint errors in data_type_transform_test.cu
* Fix compilation error
* Fix CPPLint issues in threadpool_test
* Fix CPPLInt issues in op_registry_test
* Fix CPPLint issues in operator_test
* Fix compilation error
* test
7 years ago
c08752ca43
Fix compile error
7 years ago
feaf168d95
Fiux compile error
7 years ago
b6ec45d8da
Fix CPPLint warnings in tensor_util_test
7 years ago
cb7f096da1
Fix cpplint error in mkldnn_activation ( #10105 )
7 years ago
e66f0c73af
Fix CPPLint errors in framework/details ( #10104 )
7 years ago
744ebcfa18
Fix CPPlint issues in fluid/inference ( #10075 )
7 years ago
7a993ee4f7
Merge pull request #10080 from jacquesqiao/refine-listen-and-serve-op
...
Refine listen and serve op
7 years ago
f2e400d65b
Revert "accelerate dropout ( #9902 )" ( #10082 )
...
* Revert "accelerate dropout (#9902 )"
This reverts commit 2e331c6593
7 years ago
0763ae9a1a
remove unused file
7 years ago
dc3d2dc8ff
rename grad_map to grad_to_id
7 years ago
260bf5aceb
add sync_mode
7 years ago
eb8e14c9fd
Merge pull request #10081 from chengduoZH/refine/gather_broadcast
...
Fix scope of gather and broadcast, and code clean
7 years ago
63fbdcf979
update send_recv_op_test
7 years ago
e2ace032ae
rename RunAsyncUpdate to RunAsyncLoop
7 years ago
f997c9b702
Merge branch 'refine-listen-and-serve-op' of ssh://github.com/jacquesqiao/Paddle into add-async-listen-and-serv-op
7 years ago
0f5a9cc9fc
change RunSyncUpdate to RunSyncLoop
7 years ago
df70d5f1ce
Fixed some bugs
7 years ago
0a881a1ecf
init RunAsyncUpdate
7 years ago
36083018c1
Merge branch 'refine-listen-and-serve-op' of ssh://github.com/jacquesqiao/Paddle into add-async-listen-and-serv-op
7 years ago
acd7309ea1
Merge pull request #10079 from jacquesqiao/fix-reduce_op_handle_test
...
fix reduce_op_handle_test
7 years ago
f738691777
Merge pull request #9740 from dzhwinter/memory/activation
...
"polish activation"
7 years ago
9a4ae4df79
fix scope of gather broadcast
7 years ago
d144dba4a1
simplify code
7 years ago
9c2d7df8ad
optimize code
7 years ago
8e00540778
Merge pull request #10058 from Xreki/core_fix_flush
...
Add flush of program desc to update the proto information.
7 years ago
570be39195
fix build
7 years ago
9a0ad10ff1
fix reduce_op_handle_test
7 years ago
8f7c77309d
refine listen_and_serv_op
7 years ago
cec4e6ed0d
Merge pull request #9946 from chengduoZH/feature/add_reduce_op_handle
...
Feature/add reduce op handle
7 years ago
23a21c86b1
Merge pull request #9922 from chengduoZH/feature/refine_gather_reduce
...
Refine gather and broadcast
7 years ago
1e30c41e7b
add split string
7 years ago
d002aa7abf
update
7 years ago
92b5f49ee4
Eager destruct local scope
7 years ago
a39e607798
Merge branch 'develop' of https://github.com/PaddlePaddle/Paddle into add-async-listen-and-serv-op
7 years ago
ba927b8811
Merge pull request #10060 from jacquesqiao/update-variable-response
...
VariableResponse support deserialize var into local scope
7 years ago
324ab7a39a
Fix CPPLint issues with select_op ( #10072 )
7 years ago
122141249d
Fix cpplint for print_op ( #10070 )
...
* Fix print op cpplint errors
* Remove commented code
7 years ago
8113de9425
Fix more CPPLint errors ( #10069 )
7 years ago
65b3138e98
add check
7 years ago
bb4b9af7d4
Merge pull request #10056 from typhoonzero/fix_splitbyref_macbuild
...
fix split byref op mac build
7 years ago
d060b5dfac
Registered beam search op
7 years ago
b94c518884
Implemented BeamSearchKernel
7 years ago
0ffd33d30e
VariableResponse support deserialize var into local scope
7 years ago
7ffbcbcaf0
Add flush of program desc to update the proto information.
7 years ago
ba8b0a5dc2
fix mac build
7 years ago
aa4d51b095
Merge pull request #10029 from wanghaoshuang/fix_gru_unit
...
Remove intermediate output's gradient from inputs of grad_op.
7 years ago
2e331c6593
accelerate dropout ( #9902 )
...
* accelerate dropout
* accelerate dropout
* "fix the dropout test"
* "rerun ci"
* "fix ci"
* "rerun ci"
* "fix ci"
* "fix"
* "stage"
* disable
7 years ago
881ea62bbf
Added BeamSearchOpMaker class
7 years ago
e84f353e1a
optimize
7 years ago
0b8630b9b4
Merge pull request #9897 from Yancey1989/auto_grwon_sparse_table
...
Auto-grown sparse table
7 years ago
82b192a3fd
Merge pull request #10041 from abhinavarora/cpplint_framework
...
Fix CPPLint errors in some framework files
7 years ago
88f8183576
Merge branch 'develop' into feature/refine_gather_reduce
7 years ago
1fc9fe773d
merge develop
7 years ago
035712822c
fix VisitVariable
7 years ago
216443de15
Merge branch 'develop' of https://github.com/PaddlePaddle/Paddle into fix_gru_unit
7 years ago
cbbf08aee9
Fix CPPLint errors in some framework files
7 years ago
e04c43d543
add semicolon to op registry ( #10034 )
...
* script to add semicolon
* fix typo
7 years ago
c5c7dc2e82
Fix CPPLint errors in multiclass_nms, nccl, nce, reduce and save_load_combine ( #10032 )
...
* Fix CPPLint errors in multiclass_nms, nccl, nce, reduce and save_load_combine
* Fix
7 years ago
598035f985
Fix a bug in save_inference_model and prune when the program is initailized by load_inference_model ( #10011 )
...
* Fix bug in save_inference_model and prune when the program is initialized by load_inference_program.
* Save the transpiled program instead.
7 years ago
9ca578d49e
Fix CPPLint issues in expand_op, gather_op and get_places_op ( #10000 )
7 years ago
98c12b1a08
Clean up C++ codes. ( #10022 )
...
* Privatize OpHandleBase
* Clean up a few private members
7 years ago
1a43828780
implement main logic
7 years ago
9b8ca0cfa8
"fix after merge"
7 years ago
7a517dc93e
merge develop
7 years ago
00548a1601
Remove intermediate output's gradient from inputs of grad_op.
7 years ago
d84cdb7b59
Merge pull request #9911 from tonyyang-svail/unify_op_registry
...
Unify REGISTER_OP and REGISTER_OPERATOR
7 years ago
d655417f90
Merge pull request #9956 from typhoonzero/split_byref_op
...
Split byref op
7 years ago
17212696cd
Added BeamSearchOpMaker class
7 years ago
f12b3f3617
use memcpy
7 years ago
61cb4f2fdb
"fix ci"
7 years ago
60e0d1aabd
Merge pull request #10023 from reyoung/feature/DtorOfPE
...
Correctly implement destructor of ParallelExecutor
7 years ago
28013ef9e2
Merge pull request #10019 from reyoung/feature/feed_fetch_tensor_on_cpu
...
Fix FetchTensor on CPU
7 years ago
13e7194ebd
Merge branch 'develop' of github.com:PaddlePaddle/Paddle into auto_grwon_sparse_table
7 years ago
chengduoZH
Kexin Zhao
Yao Cheng
chengduo
fengjiayi
Luo Tao
Yu Yang
gmcather
typhoonzero
Xin Pan
Yan Chunwei
Yiqun Liu
Darcy
dzhwinter
Lei Wang
Kexin Zhao
Siddharth Goyal
Yancey
Qingsheng Li
Abhinav Arora
Tomasz Patejko
yi.wu
Qiao Longfei
Yang yaming
Helin Wang
qingqing01
Yancey1989
whs
dyning
baiyf
gongweibao
Abhinav Arora
Yang Yang(Tony)
wangyang59
Yi Wang
dzhwinter