Commit Graph

14980 Commits (cbdb8a17b1e7ab27fd7429c4a94db898d9a05b24)

Author SHA1 Message Date
Yiqun Liu 112f16143b
Add an option to enable the cache of expected kernel in train phase. (#16724)
6 years ago
liuwei1031 2e07c19a9c
disable memory_optimize and inpalce strategy by default, test=develop (#16760)
6 years ago
dongdaxiang ea07eb8cd2 remove comment in data_feed.cc
6 years ago
Tao Luo 544f91deba add WITH_HIGH_LEVEL_API option, default OFF
6 years ago
guru4elephant e349a7443f
Update nccl_context.h
6 years ago
Qiao Longfei a541c25ab6 fix cpplint test=develop
6 years ago
dongdaxiang 05464e7c5c add gpu training for Executor.train_from_dataset
6 years ago
Qiao Longfei 0608f8ca56 Merge branch 'develop' of https://github.com/PaddlePaddle/Paddle into add-async_sparse_param_update_recorder
6 years ago
heqiaozhi e9d79dd5d7 ctx.device_context() to CPUPlace
6 years ago
heqiaozhi 44b226eda6 ctx.device_context() to CPUPlace
6 years ago
heqiaozhi fa6ea1e0e6 remove grad X
6 years ago
heqiaozhi 72c9aecfc3 fix doc
6 years ago
heqiaozhi 8de5dc31db add doc
6 years ago
heqiaozhi 5204fb4402 add doc
6 years ago
heqiaozhi 6e5c44d3fe add doc
6 years ago
gongweibao bf606bce8a
Fix grpc log message. (#16735)
6 years ago
Zeng Jinle 9f7b027dce
fix activation grad op desc maker (#16715)
6 years ago
lujun 9bd44b94da
Merge pull request #16561 from junjun315/move-api-to-root
6 years ago
heqiaozhi ba78446cca add continuous value model op
6 years ago
wopeizl 00279fdcc2
modify the build script for new ci test=develop (#16732)
6 years ago
liuwei1031 fdb719a1bf
avoid optimize variable used in subblock, test=develop (#16739)
6 years ago
Kaipeng Deng ed97156461
Merge pull request #16439 from heavengate/resize_scale
6 years ago
heqiaozhi 0c3c5e19d3 add continuous value model op
6 years ago
Tao Luo 1a21d08f12
Merge pull request #16725 from tensor-tang/pass/disable_seqpool
6 years ago
heqiaozhi 54dddee37e add continuous value model op
6 years ago
liuwei1031 a18ef10c87
only use the latest version variable for inplace strategy (#16736)
6 years ago
Huihuang Zheng 2146293d26 Fix op registry (#16677)
6 years ago
Tao Luo 5c364cda3c
Merge pull request #16711 from luotao1/has_attr
6 years ago
tensor-tang d6c1b5a73b disable seqpool concat pass by default saving CI time
6 years ago
baojun 1c8b34ddd2 fix training validation test=develop (#16698)
6 years ago
lujun 92c8ac8a74 merge conflict, test=develop
6 years ago
chengduo 55b15db5af
Add unit test for fuse all_reduce ops (#16699)
6 years ago
luotao1 4098ba29ed reduce hasAttr elapsed time in RunImpl
6 years ago
luotao1 f89a9c5d95 Merge branch 'develop' into has_attr
6 years ago
Tao Luo ad4a1bd13c
Merge pull request #16339 from luotao1/core_opt_choose_kernel
6 years ago
luotao1 6afc97ca6b reduce hasAttr elapsed time in RunImpl
6 years ago
Yan Xu 55e3c6949b
disable reuse port test=develop (#16704)
6 years ago
gongweibao 8b793d0efd
Fix DGC bug. (#16697)
6 years ago
Yiqun Liu 3fe8cb0dd7
Enable the runtime_context_cache pass in train phase (#16640)
6 years ago
Tao Luo 4048a2681f
Merge pull request #16687 from luotao1/reduce_inference_ci_time
6 years ago
Yan Xu 169829c83a fix win gpu test=develop (#16694)
6 years ago
guru4elephant 7d653f0aed
Merge pull request #16652 from xjqbest/dataset_merge_develop
6 years ago
xjqbest 6a57e8075a remove trainer_id in datafeed and dataset
6 years ago
tensor-tang ad45a08351
fix avx option (#16683)
6 years ago
Tao Luo d5c8d4acfe reduce all analyzer_test ci elasped time
6 years ago
luotao1 695f2db6a0 update expected_kernel_cache_pass
6 years ago
luotao1 226596a296 Merge branch 'develop' into core_opt_choose_kernel
6 years ago
Tao Luo 0d6581c574
Merge pull request #16656 from PaddlePaddle/revert-16555-model_data_cryption_link_all_lib
6 years ago
bingyanghuang 88ceda5134 MKLDNN INT8 v2 readme.md (#16515)
6 years ago
Tao Luo 503b5eb4dd
Merge pull request #16678 from luotao1/int8_cache
6 years ago
Yan Xu b4c3a6aa0b
[Imperative] implement imperative NCCLParallelContext (#16477)
6 years ago
Zeng Jinle bb143052cb
fix gc bug in conditional block (#16673)
6 years ago
xjqbest 5e5139283b fix runtime error
6 years ago
luotao1 bd636a9ea6 test_analyzer_int8 tests use default pass order
6 years ago
ruri 229dc93277
Add Pixel shuffle OP (#15782)
6 years ago
lujun b236091e89 merge confict, test=develop
6 years ago
lujun facd95e6e3 merge confict, test=develop
6 years ago
lujun 01f4f2d7e4 merge confict, test=develop
6 years ago
lujun 6173585051
Merge pull request #16634 from junjun315/my-cool-stuff
6 years ago
Chen Weihang 0b2aec14b6 Revert "Model data cryption link all lib (#16555)"
6 years ago
Yan Chunwei 044ae2497d
fix identity temporarily (#15942)
6 years ago
xjqbest 271b7147cc fix dataset bug
6 years ago
Zeng Jinle 1c526e1d1a
Fix some grad op desc makers (#16633)
6 years ago
chengduo ea2a2f778a Fix the bug of AllReduceDepPass (#16393)
6 years ago
Wojciech Uss ec2750b3c2 fix repeating passes (#16606)
6 years ago
Wojciech Uss 9b6a029666 fix dataset reading and add support for full dataset (#16559)
6 years ago
lujun 3773260edd fix load type, test=develop
6 years ago
chengduo b75a69bad6
Add Stream for fetch op handle (#16600)
6 years ago
chengduo 1342e2ea04
Fix the bug of the fast threaded executor (#16514)
6 years ago
Zeng Jinle d658244997
fix some grad op desc maker (#16581)
6 years ago
gongweibao 423bc515da
fix batch merge bug (#16601)
6 years ago
gongweibao 0342f01249
Fix dgc bug. (#16602)
6 years ago
lidanqing 2ca0de3cd4 fix preprocess script with processbar, integrity check and logs (#16608)
6 years ago
Tao Luo c797aed893
Merge pull request #16584 from luotao1/enhance_wget
6 years ago
Qiyang Min 12e36d38a5 Imperative deep-first backward process (#16605)
6 years ago
liuwei1031 bd193781df
fix the bug of reusing different types of variables in memory_optimiz… (#16547)
6 years ago
Chen Weihang c38c7c5619
Model data cryption link all lib (#16555)
6 years ago
乔龙飞 Qiao Longfei 21622ca30b
Merge pull request #16172 from jacquesqiao/add-async-ssa-graph-executor-communicator
6 years ago
Tao Luo ce18710421 enhance analyzer_tests download
6 years ago
Qiao Longfei 766666a957 add log for FLAGS_communicator_send_wait_times
6 years ago
sneaxiy 10249c0b78 Merge develop
6 years ago
zhoukunsheng 5edf4fb4fb Merge branch 'develop' of https://github.com/PaddlePaddle/Paddle into all_any
6 years ago
zhoukunsheng 47a5eec87d improve comment
6 years ago
Qiao Longfei 4031c1a7b1 fix ci build test=develop
6 years ago
zhoukunsheng 68f887aab3 Merge branch 'develop' of https://github.com/PaddlePaddle/Paddle into zeros_like
6 years ago
zhoukunsheng b2cdadd700 improve comment
6 years ago
Qiao Longfei 9861a92f6f change the return type of NewTempScope to unique ptr test=develop
6 years ago
dengkaipeng d02f4c2fd6 fix API.spec. test=develop
6 years ago
zhoukunsheng 315176c65b Merge branch 'develop' of https://github.com/PaddlePaddle/Paddle into rank
6 years ago
zhoukunsheng 8b6b626718 fix problem of API.spec mismatch
6 years ago
sneaxiy e8747d8387 fix linux ci check
6 years ago
chengduo feb1b54f9d
fix min and max bug (#16570)
6 years ago
石晓伟 5dea0bdd1b
Merge pull request #16498 from Shixiaowei02/feature/anakin-engine
6 years ago
Qiao Longfei fb6cc3a1bd follow commnet, optimize code and add comment test=develop
6 years ago
Qiao Longfei adf272bcec Merge branch 'develop' of https://github.com/PaddlePaddle/Paddle into add-async-ssa-graph-executor-communicator
6 years ago
guru4elephant 76b49f02ee
Merge pull request #16539 from guru4elephant/train_with_pipe_reader_merge_develop
6 years ago
Qiao Longfei baf02328b2 Merge branch 'develop' of https://github.com/PaddlePaddle/Paddle into add-async-ssa-graph-executor-communicator
6 years ago
qingqing01 1ebd7434d5
Add linear learning warmup method in learning rate scheduler. (#16563)
6 years ago
Qiao Longfei 9db1a9e128 change log level test=develop
6 years ago
gongweibao a61ed9782e
fix log level test=develop (#16554)
6 years ago
Wu Yi 22b02bfa62
Batch norm cudnn accurate (#16545)
6 years ago
Qiao Longfei 8342f12e31 fix set remote_prefetch test=develop
6 years ago
Qiao Longfei df45c8c538 update nce and hierarchical_sigmoid remote_prefetch
6 years ago
Qiao Longfei a1821a0449 remote remote_prefetch in embedding layer test=develop
6 years ago
zhaoyuchen2018 0b0abdbcd4
Merge pull request #16041 from zhaoyuchen2018/docrefine
6 years ago
dongdaxiang d7963e1065 infer_from_dataset API.spec
6 years ago
dongdaxiang 718ea6dbd5 fix fleet code style
6 years ago
xjqbest 782ab2e2bd add some doc
6 years ago
xjqbest 9b84e8e66b fix code style
6 years ago
xjqbest a99c8d0c29 fix client to client communication bug
6 years ago
gongweibao fea91164b7 Fix windows compilation error! (#16546)
6 years ago
dongdaxiang b35d27fa94 fix API spec
6 years ago
Kaipeng Deng 3d939d32ee
Merge pull request #16023 from heavengate/kl_div_loss
6 years ago
Kaipeng Deng 54474637ae
Merge pull request #16057 from heavengate/softmax_axis
6 years ago
Kaipeng Deng 63ac947e2f
Merge pull request #16135 from heavengate/shift
6 years ago
chengduo bb80dae7d0
Add DecoupledWeightDecay (#16427)
6 years ago
dengkaipeng 737ea825ac fix doc. test=develop
6 years ago
dongdaxiang 3829eac27b fix API spec about infer_from_dataset
6 years ago
Shixiaowei02 7b9fc71076 update tensorrt subgraph_util test=develop
6 years ago
dengkaipeng 0f7411a1ae round down for scale. test=develop
6 years ago
dongdaxiang 87027a2eef fix API.spec problem and executor's docstring
6 years ago
zhoukunsheng 5284213942 Merge branch 'develop' of https://github.com/PaddlePaddle/Paddle into rsqrt
6 years ago
zhoukunsheng 91ba75000c fix type conversion problem in rsqrt functor
6 years ago
Tao Luo 8f7b5883b8
Merge pull request #16529 from lidanqing-intel/lidanqing/preprocess-data
6 years ago
Tao Luo 5b24002389
Merge pull request #16399 from sfraczek/sfraczek/analyzer_int8_resnet50_test
6 years ago
sneaxiy 8c869a865d update develop ops
6 years ago
sneaxiy 33473890f3 Merge develop
6 years ago
dongdaxiang ade9337486 fix API.spec
6 years ago
liuwei1031 278debab71
fix comments of 16410, test=develop (#16499)
6 years ago
Wojciech Uss 2498395132 remove profiling from int8 test
6 years ago
Zhaolong Xing 3e6aa498d6
Merge pull request #16526 from NHZlX/refine_trt_anakin
6 years ago
nhzlx 7cde2d9e84 fix trt engine test error.
6 years ago
zhoukunsheng 3c4f5f0368 Merge branch 'develop' of https://github.com/PaddlePaddle/Paddle into linspace
6 years ago
zhoukunsheng ead3c0a8fc update api.spec
6 years ago
dongdaxiang 720647e17f rebase current develop and fix conflict
6 years ago
zhoukunsheng 2336d5ca5d Merge branch 'develop' of https://github.com/PaddlePaddle/Paddle into rank
6 years ago
dongdaxiang 3a79be6eb3 refine API spec
6 years ago
dongdaxiang 98dda08a85 fix pull sparse slow problem
6 years ago
dongdaxiang 93c3c7f9b3 fix dataset testcase problem
6 years ago
dongdaxiang d739bab844 fix async_executor problem and remove some unnecessary testcase, fix trainer_desc import problem
6 years ago
dongdaxiang 241d8808be add timer to distributed executor
6 years ago
dongdaxiang 3c73859eec add trainer_desc.proto to distributed executor
6 years ago
dongdaxiang 60b7bf6fa6 add infer_from_dataset for inference
6 years ago
xjqbest 030c7e7e9d fix FillSparseValue error
6 years ago
dongdaxiang 88880d9b69 fix import trainer_desc_pb2 error
6 years ago
dongdaxiang 0030eb2a61 fix distributed building
6 years ago
dongdaxiang ed31874397 undefine rand_r()
6 years ago
dongdaxiang f7e4813804 add WIN32 for rand_r and usleep
6 years ago
dongdaxiang cedbc161da add more _LINUX maroc on data_feed.cc for mac and window compile
6 years ago
dongdaxiang c5980c3566 add _LINUX macro
6 years ago
dongdaxiang 433301fbc2 remove glog in shell.h
6 years ago
dongdaxiang 9e51ad4a65 fix io and fs compile on mac
6 years ago
dongdaxiang 6eca88ac76 fix io and fs compile on mac
6 years ago
dongdaxiang 2708108a08 fix fleet_wrapper compile on windows
6 years ago
dongdaxiang 4ce35815fb fix windows GLOG problem
6 years ago
dongdaxiang e3107a6ae0 fix windows compile problem
6 years ago
dongdaxiang 398004ece0 disable sys/wait.h to fix windows compile problem, include scope in lodtensor_printer
6 years ago
dongdaxiang d4514949bf remove local random engine in fleet with rand_r()
6 years ago
dongdaxiang e82969eeb0 remove getdelim in windows
6 years ago
dongdaxiang 45eb6f0765 run pre-commit check files and fix code style problem
6 years ago
dongdaxiang d87ba58c14 refine document of python API, make device_worker and trainer's API private
6 years ago
dongdaxiang 5687f234bf fix trainer_desc.proto error
6 years ago
dongdaxiang b95b80bc76 add doc string for executor and update API.spec
6 years ago
dongdaxiang 6be9f719e2 make string_helper dependency work
6 years ago
xjqbest e95cafd9a7 fix code style & add dataset testcase
6 years ago
dongdaxiang 39362a8415 move root_scope->DropKids() into Finalize() so that we do not have to drop all the kids
6 years ago
dongdaxiang ba15d6b164 move root_scope->DropKids() into Finalize() so that we do not have to drop all the kids
6 years ago
xjqbest be74de2c61 fix code style & fix register bug & add release_memory
6 years ago
dongdaxiang a0b59773af fix code style
6 years ago
dongdaxiang f39b323ed7 remove trainer_library in CMakeLists
6 years ago
dongdaxiang 365be5d559 support win32 flag in io.cc shell.cc, fix code style problem in fleet_wrapper, fix lodtensor_printer_test problem
6 years ago
dongdaxiang dc8cf36e4b add more example on datagenerator
6 years ago
dongdaxiang 6bf796df14 refine print fetch list
6 years ago
xjqbest 589467f24c fix bug
6 years ago
xjqbest b7940c2918 fix bug of gen_worker_desc and set_filelist, add some doc
6 years ago
dongdaxiang 68d7bf3de5 add fetch var function
6 years ago
xjqbest a34fe6248f add some doc
6 years ago
xujiaqi01 f5c6a14b54 fix runtime error
6 years ago
xujiaqi01 a5b1a0e12b support multi dataset && add init model && fix bug
6 years ago
dongdaxiang 3c65cc1bbd add document for role_maker and fleet parameter, data_generator
6 years ago
dongdaxiang f6c9232a3d fix dataset float32 type problem
6 years ago
dongdaxiang 73b1f396d7 add data_generator into paddle.fluid.incubate.data_generator, add op run log in hogwild_device_worker and downpour_device_worker
6 years ago
dongdaxiang 73544e8b8d add training speed log
6 years ago
dongdaxiang 9419de521f add IO percent for multi_trainer
6 years ago
dongdaxiang 6af697adb0 add trainfileswithprofiler for downpour worker
6 years ago
dongdaxiang 2644b88685 add comment for MPI Symetric role maker
6 years ago
dongdaxiang cf45c54340 add distributed optimizer factory
6 years ago
dongdaxiang b7a202aa38 add distributed optimizer factory
6 years ago
xujiaqi01 70a5d4f797 fix error
6 years ago
xujiaqi01 d25389fefd add some log && fix error
6 years ago
dongdaxiang f612877797 add incubate for unified API
6 years ago
dongdaxiang 317eb0aad3 add incubate for unified API
6 years ago
xujiaqi01 39449ba0b9 fix bug && add DestroyReaders in trainer
6 years ago
dongdaxiang e657c127a8 hide opt_info in distirbuted optimizer
6 years ago
xujiaqi01 ecfc7df913 add dataset factory && fix style
6 years ago
dongdaxiang 328f11b8b6 refactor downpour optimization
6 years ago
xujiaqi01 3cea00bd52 store memory data in Dataset && fix bug
6 years ago
dongdaxiang ff87698a44 refactor downpour optimization
6 years ago
dongdaxiang b66f0074b6 fix data reading bugs in api, add VLOG(3) log for setup
6 years ago
dongdaxiang b415ec27e8 make Dataset* as an argument
6 years ago