Commit Graph

851 Commits (31464f3425da4fee13d88859a0e0f59f29fbad48)

Author SHA1 Message Date
JiayiFeng c0257f0a5b Add comments
7 years ago
JiayiFeng 5aa440fd7a Add move constructor for Item
7 years ago
Yi Wang 1bbbc4e76f Merge branch 'develop' of http://github.com/paddlepaddle/paddle into fix_cpplint_errors_operators_detail
7 years ago
qiaolongfei fdecae5fc5 Merge branch 'develop' of https://github.com/PaddlePaddle/Paddle into prefetch_on_server
7 years ago
qiaolongfei 3a5bce775e try to complete
7 years ago
Yi Wang 767f453ab8
Add cpplint pre-commit hook (#9511)
7 years ago
JiayiFeng a469666e42 fix compile errors
7 years ago
fengjiayi 2f856769b9 Merge branch 'develop' of https://github.com/PaddlePaddle/Paddle into refine_double_buffer_fix
7 years ago
fengjiayi 55e4b89f14 remove local_buffer_
7 years ago
fengjiayi 7bb18433fd refine code
7 years ago
Qiao Longfei 63cd5fb0b1
Merge pull request #9523 from jacquesqiao/fix-test_send_recv
7 years ago
Tao Luo 527e6585d1
Merge pull request #9528 from tpatejko/tpatejko/mkldnn-lrn-fix-is_test_failure
7 years ago
Yancey1989 c3580eae46 Add prefetch interface on server side
7 years ago
Yu Yang 53fa7cb9cc Add local cache of double buffer reader
7 years ago
Tao Luo 7102eb2efd
Merge pull request #9531 from luotao1/fix_profiler_test
7 years ago
chengduoZH ffa63974b9 compare the performance of unpinned memory and pinned memory
7 years ago
Tomasz Patejko b9874251c6 Plain LRN op throws an exception when is_test is set in backward pass
7 years ago
Luo Tao 5baa529e0e fix compiler error of profiler_test in ONLY_CPU mode
7 years ago
fengjiayi 95658767eb
Merge pull request #9428 from JiayiFeng/kernel_of_increment_op
7 years ago
typhoonzero 52439d9f1d Merge branch 'fix-test_send_recv' of https://github.com/jacquesqiao/Paddle into fix_server_shutdown
7 years ago
typhoonzero f6de248323 fix server shutdown
7 years ago
chengduo 81d93514d6
Merge pull request #9522 from chengduoZH/feature/refine_parallel_exe
7 years ago
Qiao Longfei 23bab34ca3
Fix data transform when inplace (#9450)
7 years ago
chengduoZH 60d0a0594e refine parallel
7 years ago
Yancey 374f1ca3b7 Fix dist error with lr decay layer (#9489)
7 years ago
Qiao Longfei f0af1398b8
add prefetch_op (#9495)
7 years ago
Yu Yang fa21436d0d
Merge pull request #9080 from reyoung/cpp_parallel_executor
7 years ago
Yi Wang c1c5e166d1 Fix cpplint errors
7 years ago
Yi Wang 64242c5d71 Rename test_serde into serde_test
7 years ago
Abhinav Arora 5f9da86ba5
Fix the order of reads and write from buffered channel (#9423)
7 years ago
Yang Yu af230d9bef Merge branch 'develop' of https://github.com/PaddlePaddle/Paddle into cpp_parallel_executor
7 years ago
qiaolongfei e727cdb62d fix block num
7 years ago
fengjiayi 1a4b0d63e4
Merge pull request #9352 from JiayiFeng/doc_update_reader_doc
7 years ago
dzhwinter 8425c2c859
Speed/sequence op1 (#9217)
7 years ago
guosheng 5b8bb34470 Refine reshape_op by following comments.
7 years ago
fengjiayi 1e4f442a84 fix a compile error
7 years ago
武毅 d21ab2e2ba
Merge pull request #9448 from typhoonzero/fix_dist_slr_height
7 years ago
chengduo 24100e1fb8
Merge pull request #9449 from chengduoZH/feature/add_cos
7 years ago
fengjiayi 869ef01d66 Merge branch 'develop' of https://github.com/PaddlePaddle/Paddle into kernel_of_increment_op
7 years ago
JiayiFeng 52574733a6 Add KernelType switch for IncrementOp kernel
7 years ago
typhoonzero 96192a85ec Merge branch 'develop' of https://github.com/PaddlePaddle/Paddle into fix_dist_slr_height
7 years ago
Abhinav Arora f5da16e51b
Disabling channel test to debug issue (#9491)
7 years ago
JiayiFeng 0ac43217ce check whether scalar condition var is on CPU before using
7 years ago
typhoonzero 450be963fe fix sparse errors
7 years ago
chengduoZH bdda08d9f2 add sin
7 years ago
JiayiFeng 01c5ca7364 fix bugs
7 years ago
Yu Yang e868950e5f Add comments
7 years ago
JiayiFeng 917b205c1c Merge branch 'develop' of https://github.com/PaddlePaddle/Paddle into kernel_of_increment_op
7 years ago
chengduoZH 2e577379ca add cos
7 years ago
Tao Luo 857a8997de
Merge pull request #9384 from luotao1/removeVar
7 years ago
Yu Yang 38b53b37b4 Remove Pop method
7 years ago
Yu Yang ce2f096372 Merge branch 'cpp_parallel_executor' of github.com:reyoung/Paddle into cpp_parallel_executor
7 years ago
Yu Yang 7da1ea07a2 Use PopAll
7 years ago
fengjiayi 802dcd676e remove CPU restrict in While_op
7 years ago
Yang Yu b0775588c0 Merge branch 'develop' of https://github.com/PaddlePaddle/Paddle into cpp_parallel_executor
7 years ago
typhoonzero 9a9d67dac2 fix dist train selected rows height missing
7 years ago
Yu Yang 084cdd1f4f Rename code
7 years ago
chengduoZH 58a9f9f781 set the max size of cudapinned memory
7 years ago
fengjiayi 6dfc33c226 fix compile errors
7 years ago
fengjiayi e9370fe59f fix compile bugs
7 years ago
fengjiayi 0ce558f19e kernels of increment op
7 years ago
guosheng 4bfbc59122 Merge branch 'develop' of https://github.com/PaddlePaddle/paddle into enhance-ReshapeOp
7 years ago
guosheng c078ed4608 Enhance reshape_op by adding Input(Shape)
7 years ago
yi.wu cc1c6afbbf fix slr serde
7 years ago
typhoonzero 094d509689 fix slr deser
7 years ago
typhoonzero 587781153e fix slr deser
7 years ago
Luo Tao 7f4012247e adjust remove rule for variables
7 years ago
Yu Yang 201f79d039 Use Extend method
7 years ago
Yu Yang dcf7bd2d92 Add initP2P
7 years ago
Yu Yang 50f71f5005 Using blocking queue
7 years ago
qingqing01 25317bd312
Make the first device share data with the global scope in parallel_do_op. (#9398)
7 years ago
Yu Yang 7dcb217e31 Refine allreduce op
7 years ago
Yu Yang c0c2e15920 NCCL AllReduce
7 years ago
Yu Yang 3f88fad08c Fix merge op
7 years ago
Yu Yang 5b92dd4026 Remove dev sync
7 years ago
Yu Yang 52dd8ff09a Force sync dev
7 years ago
Yu Yang dfb8680018 Early drop fetch op
7 years ago
Yu Yang 9af870854e Use heap variables
7 years ago
Yu Yang 222763296f Change fetch op
7 years ago
Yu Yang 76570c2e96 Wait fetch op
7 years ago
Yu Yang b6ca3711b4 Get error
7 years ago
Yu Yang 55e2cc3d87 FetchOp Force sync
7 years ago
Yu Yang 5a02739ce9 Throw error
7 years ago
Yu Yang f385228f05 Add Paddle Enforce
7 years ago
Yu Yang 833e522d16 Enhance drop kids
7 years ago
Yu Yang aba46f077b Disable P2P
7 years ago
gongweibao e0b5691e41
Add drop_out_op unit test (#9364)
7 years ago
chengduoZH ab601c19c3 Add CUDAPinnedPlace
7 years ago
Tao Luo 1b67bc022c
Merge pull request #9329 from tpatejko/tpatejko/mkldnn-lrn
7 years ago
Abhinav Arora 65534c4762
Fluid channels should match the semantics of Go Channels (#9265)
7 years ago
chengduoZH 158d6c4d19 add unit test
7 years ago
Luo Tao ccfec1bcb1 remove vars when remove ops
7 years ago
chengduoZH 18eb77303d add CUDAPinnedPlace
7 years ago
Qiao Longfei f3dc3112cc
add split ids op (#9370)
7 years ago
chengduo 2e4a398638
Merge pull request #9216 from chengduoZH/feature/add_pinned_memory
7 years ago
Tao Luo c858f48979
Merge pull request #8887 from luotao1/infer_mkl
7 years ago
typhoonzero 1ab4fcb5e7 prepare pserver executor
7 years ago
chengduoZH 9e99446e25 Add note for cudaMallocHost
7 years ago
chengduoZH a0e2cf03e4 Merge branch 'develop' of https://github.com/PaddlePaddle/Paddle into feature/add_pinned_memory
7 years ago
Yu Yang 9dd64d83f3 WMT Model
7 years ago
chengduoZH 39004080f4 replace use_pinned with is_pinned
7 years ago
Yu Yang cb40c33137 Update unittest
7 years ago
Yu Yang ee97687f69 Fix compile
7 years ago
Yu Yang 3aa2a8ffcf Follow comments
7 years ago
Yu Yang 02aaecca35 Fix CPU compile
7 years ago
Yu Yang 54bd17fe7b Complete Flowers
7 years ago
Xin Pan 3941c2ddec
Merge pull request #9355 from panyx0718/layer_norm
7 years ago
Luo Tao 6332bd1ed8 Merge branch 'develop' into infer_mkl
7 years ago
Yu Yang 50e7e25db3 Merge branch 'develop' of https://github.com/PaddlePaddle/Paddle into cpp_parallel_executor
7 years ago
Yu Yang 5c7a523326 Add Graphviz output
7 years ago
Qiao Longfei 4f522fa8d5
fix compile send_op on mac (#9360)
7 years ago
Yancey 1b0a17f415
Merge pull request #9303 from Yancey1989/split_send_op
7 years ago
Yancey1989 ebbb428db9 fix ci
7 years ago
Tao Luo cb3bbbd5c6
Merge pull request #9081 from kbinias/kbinias/mkldnn-activations
7 years ago
Qiao Longfei 8ccc61f334
support empty tensor (#9338)
7 years ago
chengduo 4a92e89623
Merge pull request #9337 from chengduoZH/feature/fix_concat
7 years ago
武毅 12856c5f69
Merge pull request #9325 from dzhwinter/fix/dropout1
7 years ago
chengduoZH aca9180a76 Merge branch 'develop' of https://github.com/PaddlePaddle/Paddle into feature/fix_concat
7 years ago
Xin Pan 1a4be55a47 Pass cpu build
7 years ago
Xin Pan 904fa05f46 Improve layer_norm speed
7 years ago
Yancey1989 79af7cc9d3 Merge branch 'develop' of github.com:PaddlePaddle/Paddle into split_send_op
7 years ago
Yancey1989 081b782434 update by comment
7 years ago
fengjiayi dd532e2086 refine MultiPassReader's doc string
7 years ago
gongweibao cffe1a9112
Profiler can get elapsed time of `sendop` (#9345)
7 years ago
Darcy 8090eb6272 added proto_desc to device_tracer's dep list (#9342)
7 years ago
Yu Yang edfd741e3a Add simple python wrapper for ParallelExecutor
7 years ago
Yu Yang a7b0d5bd26 Clean code
7 years ago
Yu Yang e3144393e3 Extract Executors to indie modules
7 years ago
Yu Yang c70b60dd70 Make executor steal graph inside
7 years ago
Yu Yang 4c3361cda8 Extract GraphExecutor
7 years ago
Yu Yang b123e43bf9 extract multi devices graph builder
7 years ago
Krzysztof Binias d8bd436fc1 Fixed tests
7 years ago
Krzysztof Binias a64b312e3a Correcting for PR comments
7 years ago
Krzysztof Binias 4466f0bec8 MKLDNN Relu Tanh Sqrt Abs activations added
7 years ago
chengduoZH 750aff10ce code refine
7 years ago
chengduoZH 043f47b27f fix concat op
7 years ago
yi.wu bb815d4364 update
7 years ago
yi.wu a9a228ad8d fix dist compile
7 years ago
Luo Tao ae820a34bc Merge branch 'develop' into infer_mkl
7 years ago
Varun Arora 76ae540f8e
Move Select to concurrency.py; incorporate outputs (#9136)
7 years ago
武毅 9c35b0dc1b
Merge pull request #9287 from typhoonzero/pserver_prepare_before_run
7 years ago
Tomasz Patejko 14ba67c0ef Function for running MKLDNN primitive added. Unittest added for is_test attribute
7 years ago
Tao Luo e027eb40d7
Merge pull request #9123 from tpatejko/tpatejko/mkldnn-lrn
7 years ago
dzhwinter e33af2414b "fast hack"
7 years ago
typhoonzero 9367f11eb7 Merge branch 'develop' of https://github.com/PaddlePaddle/Paddle into pserver_prepare_before_run
7 years ago
Yancey ee7f1ecd7c
Fix dist compile error (#9320)
7 years ago
Tao Luo 9126e626fc
Merge pull request #9165 from ROCmSoftwarePlatform/amd_cmake_01
7 years ago
guosheng b7e83d2467 Merge branch 'develop' of https://github.com/PaddlePaddle/paddle into enhance-ReshapeOp
7 years ago
qingqing01 8f8728635a
Fix bug for backward tanspiler when using parallel_do operator. (#9282)
7 years ago
typhoonzero a88cc46221 update
7 years ago
typhoonzero 972a102b92 Merge branch 'develop' of https://github.com/PaddlePaddle/Paddle into pserver_prepare_before_run
7 years ago
typhoonzero 5e6276edc1 fix transpiler bug
7 years ago
dzhwinter 13f1050ab0
"fix mixed_vector bug" (#9319)
7 years ago
Liu Yiqun 5419da6e7a Fix bug caused by block_id.
7 years ago
Yu Yang 9e3e424ecb
Merge pull request #9299 from reyoung/feature/refactor_batch_norm
7 years ago
Liu Yiqun 987a32dac3 Merge branch 'develop' into core_inference_fix_run
7 years ago
Yu Yang dd73d18bb7 Extract SSAGraph
7 years ago
sabreshao e0ac6bc436 CMake refine for HIP support.
7 years ago
gongweibao 990d6396fe
Reuduce memory copy when communication between trainer and pserver. (#9271)
7 years ago
whs b594251f89
Merge pull request #9082 from wanghaoshuang/average_model
7 years ago
Yu Yang 1d8fe2a220 Enhance device context pool (#9293)
7 years ago
Kexin Zhao 64c5c8f8b0
Merge pull request #9269 from kexinzhao/softmax_cudnn_fp16
7 years ago
Kexin Zhao b9e6364e3c
Merge pull request #9267 from kexinzhao/new_relu_fp16
7 years ago
Kexin Zhao 4eaa789730 resolve conflict
7 years ago
Tomasz Patejko 72cc64e40e Device blobs are created only in training. Added testing attribute
7 years ago
tensor-tang 7260e3a443
Merge pull request #9214 from jczaja/prv-softmax-mkldnn-operator-PR
7 years ago
Yu Yang 79989c9025 Add SSA builder
7 years ago
Yancey1989 2a4221ac07 split send op to send_vars and send_barrier
7 years ago
Yu Yang 0760aaf440 Shrink batch_norm_grad's inputs
7 years ago
guosheng 454b0a96be Remove the extra call of ValidateShape in ReshapeKernel
7 years ago
guosheng 437f7a3279 Resolve conflict according to the latest code
7 years ago
Jacek Czaja 3b95b55f07 - Softmax MKLDNN primitive integration
7 years ago
guosheng eb12cbe764 Refine reshape_op infershape
7 years ago
Yu Yang 64d7a30271 Extract SSAGraph
7 years ago
Yu Yang 8dec4ad7a1 Use int not Place for vars
7 years ago
Yu Yang 3181501013 Rerange code
7 years ago
Yu Yang f28ae6e4b1 Reorganize Code
7 years ago
Yu Yang 5c333e4143 Add dctor for dev_ctx
7 years ago
Liu Yiqun 0968753454 Enable the test of not creating variables every time.
7 years ago
Yu Yang 15f5f10ed5 AddInput/AddOutput for OpHandle
7 years ago
Yu Yang 5368e50d84 Reorganize code
7 years ago
typhoonzero 1eec926124 updates
7 years ago
Yu Yang fe7ed285d1 Extract NCCLCtxMap
7 years ago
typhoonzero e9d815e32b prepare and create op before run
7 years ago
Kexin Zhao ed2bc194c5
Merge pull request #9176 from kexinzhao/batch_norm_fp16
7 years ago
fengjiayi cd07c0f021
Merge pull request #9259 from JiayiFeng/dev_MultiEpochReader
7 years ago
Yu Yang 6ebc6bf533 ReorganizeCode
7 years ago
Yiqun Liu 7bb4ea9c13
Add an argument in Executor.Run to allow users to choose whether to create and destroy variables every time. (#9242)
7 years ago
Yu Yang a478a11e0b NCCL Guard for bcast
7 years ago
Yu Yang f2685bed81 Clean code
7 years ago
Yu Yang 41ad632341 Add NCCL Group Guard
7 years ago
Yu Yang 99fe83a020 Move nccl helper
7 years ago
Yu Yang 90f980167d Do not wait computation stream
7 years ago
Yu Yang 7ac969b88c Debug
7 years ago
fengjiayi 809530f418 Merge branch 'develop' of https://github.com/PaddlePaddle/Paddle into dev_MultiEpochReader
7 years ago
fengjiayi 7c041e48f4
Merge pull request #9182 from JiayiFeng/dev_MultipleReader
7 years ago
fengjiayi e4bd63d0e1
Merge pull request #9240 from JiayiFeng/fix_bug_in_recordio
7 years ago
typhoonzero 18461d0935 wip
7 years ago
wanghaoshuang edb4e29ab7 Merge branch 'develop' of https://github.com/PaddlePaddle/Paddle into average_model
7 years ago
Kexin Zhao b7801b9fcb small fix
7 years ago