18401 Commits (a373aa76451b83a9eeb7617f92bdc9c1ea6e9ff6)

Author	SHA1	Message	Date
ShenLiang	dae3e1f337	Solve inconsistent order in each card in dynamic graph (#30931)	4 years ago
WangXi	14d039e4a1	Fix the problem that the number of ops executed by xpu is wrong (#30961)	4 years ago
Chen Weihang	010f2caa23	try to fix reader and signal test failed (#30960)	4 years ago
Adam Osewski	3ba69809bf	Fix LayerNorm tester for gcc4.8 (#30962)	4 years ago
Qi Li	93c1d9e761	[ROCM] update fluid platform for rocm39 (part3), test=develop (#30913)	4 years ago
QingshuChen	15297a065c	fix depends of kunlun bkcl (#30945)	4 years ago
liym27	97f7a70c01	Add error message for slice op(#30851)	4 years ago
liuyuhui	87197f8c2e	[kunlun]fix sync in multi kunlun xpu dygraph training. (#30943)	4 years ago
石晓伟	99bd16eb4e	bug fix of xpu lite engine, test=develop (#30918) ... * bug fix of xpu lite engine, test=develop * xpu zero copy tensor, test=develop * revert paddle/fluid/inference/tests/api/CMakeLists.txt	4 years ago
tianshuo78520a	2e93233899	Add WITH_XPU_BKCL in Kunlun-CI (#30919)	4 years ago
Qi Li	34f1628ce8	[ROCM] update fluid platform for rocm39 (part2), test=develop (#30774)	4 years ago
Jacek Czaja	9e527d9956	[oneDNN] Added basic changes for elementwise_add_grad bf16 (#30925)	4 years ago
Chengmo	c98f144fbc	add truncated gaussian random (#30922) ... add truncated gaussian random	4 years ago
liuyuhui	4a8b8b4547	[Kunlun] add gen_bkcl_id_op, support multi XPU cards training using multiprocess (#30858)	4 years ago
liym27	39f41cb47f	Performance optimization for dynamic setitem: Call op set_value to speed up because the original call to TensorToPyArray will introduce unnecessary data copy. (#30817)	4 years ago
liuyuhui	bef46ccfc8	[Kunlun]fix include files of gen_comm_id_helper.cc (#30917)	4 years ago
wanghuancoder	aab3a3012e	add include for heterbox_trainer.cc, develop=test (#30910)	4 years ago
taixiurong	24873f4f77	dyngraph (#30892)	4 years ago
Adam Osewski	092a2b1413	More UT for LayerNormFuse pass (#30891) ... * Additionally change to not throw error from inside pass.	4 years ago
tianshuo78520a	a80fe67f84	Change cmake/third_party files for CI (#30833)	4 years ago
Jacek Czaja	abfa822650	[oneDNN]Extended adaptive pooling support for oneDNN pool kernel (#30757)	4 years ago
joanna.wozna.intel	73cdea01d4	Add bf16 fast performance verification (#30551) ... * Update Xbyak and add bf16 fast performance verification * Fix formating * Change LOG message * Trigger an update of a new tag	4 years ago
Shang Zhizhou	e6095bc2ce	fix split trt plugin initialize (#30875) ... * fix split trt plugin initialize * update	4 years ago
WangXi	6e3856d3fb	fix xpu dygraph place (#30868)	4 years ago
wanghuancoder	35c5b23f68	use iwyu clean include second time, test=develop (#30829) ... * use iwyu clean include second time, test=develop	4 years ago
cucuzg	ac2e2e6b7f	add clip_by_norm on kunlun, *test=kunlun (#30862)	4 years ago
wawltor	b7560a59ab	fix the broadcast for the large second input (#30818) ... fix the broadcast for the large second input	4 years ago
JamesLim	6e1e036a75	Implement cuda kernel for index_sample. (#30380)	4 years ago
AshburnLee	666efc2336	Call new cudnn batch norm API regardless of data type and data layout (#30157)	4 years ago
QingshuChen	5c8455d6ea	try again if kunlun memory malloc failed (#30855) ... * try again if kunlun memory malloc failed * minor	4 years ago
石晓伟	2ac4143b6c	support xpu with analysis predictor, test=develop (#30832) ... * support xpu inference with analysis predictor, test=develop * merge the cmake of the xpu toolchain, test=develop * add c-apis, test=develop * fix a bug in extern_xpu, test=develop	4 years ago
liuyuhui	2cb55eff57	fix WITH_XPU_BKCL in CMakeLists.txt (#30854)	4 years ago
Adam Osewski	4f066e316e	Layer normalization fuse pass. (#30721)	4 years ago
WangXi	b1026f64af	【kunlun】dygraph supports multi xpu card training (#30671)	4 years ago
joanna.wozna.intel	04532b8a83	Update Xbyak to v5.81 (#30809)	4 years ago
Shang Zhizhou	b909450994	fix trt plugin clone and initialize bugs in TRT7.1+ (#30709) ... * fix trt plugin clone and initialize bugs * fix unit test error * enable trt in ci py3 * update unittest timeout	4 years ago
Wilber	b08ae368bb	ci compilation depends on a stable release (#30755) ... * update lite tag * disable ut	4 years ago
Thunderbrook	cb66c53c2d	dump to cpu (#30750) ... * dump to cpu * format * format * format	4 years ago
Chengmo	d3fac0ea85	fix int64 bug (#30780) ... fix push sparse int64 bug	4 years ago
Qi Li	69875dc42c	[ROCM] update fluid memory for rocm35 (part1), test=develop (#30758)	4 years ago
QingshuChen	c35a9880f9	fix malloc L3 failed bug for kunlun (#30745) ... * fix malloc L3 failed bug for kunlun * minor	4 years ago
WangXi	31ed9c9eed	Fleet distributed strategy support pure fp16 (#30754)	4 years ago
Zhen Wang	53d01afed6	Fix the nan bug when passing all zero values into clip_by_norm_op. (#30777)	4 years ago
ShenLiang	3858f458ea	rm Singleton of reducer (#30775)	4 years ago
Qi Li	f89da4ab45	[ROCM] update fluid platform for rocm35 (part1), test=develop (#30639) ... * [ROCM] update fluid platform for rocm35 (part1), test=develop * address review comments, test=develop	4 years ago
Wojciech Uss	fc00240575	A fix for oneDNN matmul kernel. Fixes issue #30309 (#30723)	4 years ago
lidanqing	46989e889b	Fix python3 incompatibility issues (#30698) ... * solve python3 incompatibility issues * update checksum	4 years ago
alncat	5b59499e57	fixed compilation error on gcc 4.8.x due to the usage of isfinite (#30733)	4 years ago
Chengmo	78d37c3f75	【Paddle.Fleet】Fix brpc get hostname (#30703) ... * fix Brpc get hostname	4 years ago
taixiurong	caf3680bbc	fix bugs in transformer predict in xpu place (#30730) ... * transformer predict * trans bug fix	4 years ago
jakpiase	f8da5536ed	REUPLOAD Added vanilla LSTM and LSTM with peepholes oneDNN fp32 kernel (#30719) ... * added external reorder to profiler * resolved conflict * added enable_static * initial version of lstm, not working yet * added lstm to operators.cmake * added vanilla lstm mkldnn op * added peephole weights integration * minor changes * added formatting * added fusion_lstm_mkldnn to static_whitelist * added formatting * removed comment * moved use_peepholes attribute inside is_cached block * reverted wrong changes * minor formatting change * minor changes * changed stream handling * minor change * added datatype to GetExpectedKernelType() * added reading stream from TLS	4 years ago
liuyuhui	67abfc1588	[Kunlun] fix dead lock for exec_op_count_ (#30718)	4 years ago
alncat	5ace20fc3f	modified conv+bn fuse pass to fix wrong mask in mask rcnn (#30704)	4 years ago
Tao Luo	824a79d383	Revert "Added vanilla LSTM and LSTM with peepholes oneDNN fp32 kernel (#30661)" (#30708) ... This reverts commit d834f4e6e8.	4 years ago
lilong12	7fbc68a2c0	update, test=develop (#30692)	4 years ago
jakpiase	d834f4e6e8	Added vanilla LSTM and LSTM with peepholes oneDNN fp32 kernel (#30661) ... * added external reorder to profiler * resolved conflict * added enable_static * initial version of lstm, not working yet * added lstm to operators.cmake * added vanilla lstm mkldnn op * added peephole weights integration * minor changes * added formatting * added fusion_lstm_mkldnn to static_whitelist * added formatting * removed comment * moved use_peepholes attribute inside is_cached block * reverted wrong changes * minor formatting change * minor changes	4 years ago
arlesniak	5bf25d1e8b	More precise mkldnn kernel rules in GetExpectedKernelType (#29840) ... * More precise mkldnn kernel choice in GetExpectedKernelType * Fixes after review * Refresh develop for CI * CI experiment * get back from CI exper	4 years ago
Jacek Czaja	173660be7b	[oneDNN] Cache oneDNN stream not to recreate in each oneDNN op (#30358)	4 years ago
Shang Zhizhou	ae0f88a988	add DLA support：C++&&Python api (#30165) ... * add dla * add dla done * add python api Co-authored-by: shangzhizhou <root@szth-rp-fanyi-opera49.szth.baidu.com>	4 years ago
chentianyu03	fb7fbc7a5d	fix abs bug and add abs test case (#30637) ... * add abs test case * use std::abs to fix abs bug * fix the abs bug * fix abs bug	4 years ago
ShenLiang	9514b4aa5f	Fix scatter grad bug (#30604)	4 years ago
Pei Yang	cf9bdb9404	extend trt ut timeout threshold (#30537)	4 years ago
Thunderbrook	1bebc09253	solve build gpu task core (#30626) ... * build gpu task core * format	4 years ago
石晓伟	33bf6eb753	revert external gflags, test=develop (#30623)	4 years ago
Jacek Czaja	dfdb0359ea	- Disabling oneDNN inplace pass (#30588)	4 years ago
TTerror	10271ddfc4	support reduce_max op on kunlun (#30581) ... * support reduce_max op on kunlun * support reduce_max op on kunlun * support reduce_max op on kunlun * support reduce_max op on kunlun	4 years ago
QingshuChen	5013c67644	fix softmax bug for multi_card in kunlun (#30600)	4 years ago
wuhuanzhou	7e671c07b6	optimize unity build (#30195) ... * optimize unity build, test=develop * fix code style error, test=develop * fix code style error and test /MP settings, test=develop	4 years ago
liuyuhui	e5b0d9e1fc	[Kunlun] Add condition_variable and notify() in BindThreadedSSAGraphExecutor (#30586)	4 years ago
Zhou Wei	9674e440e2	optimize windows CI, clear tp cache,polish code,improve level of msvc log (#30579)	4 years ago
wanghuancoder	90773473a0	use nvtx push pop in timeline (#30567) ... * delete empty line of pybing.cc, test=develop * use nvtx push pop in timeline, test=develop * change year, test=develop * add #ifdef PADDLE_WITH_CUDA, test=develop * add #ifndef WIN32, test=develop * is_pushed to is_pushed_, test=develop	4 years ago
chentianyu03	358106fcb0	make abs op support complex types (#30375) ... * rewrite abs op * rewrite abs op and remove abs in activation * remove abs register in old codes * fix abs_grad type error * fix abs double_grad output name error * modify abs_grad, abs_grad_grad functor for windows building * format code style * fix the bug of result is nan when the divisor is zero * add missing abs attr and add abs for float16	4 years ago
Wilber	2d5758c456	update. (#30585)	4 years ago
Tao Luo	9dd71c74df	disable test_analyzer_detect (#30541)	4 years ago
tangwei12	c9e78a22c5	add trainers for pserver (#30523) ... * add trainers for pserver Change-Id: I1a75793ec81ce126d07f4c47cae09b95d530bbc8	4 years ago
wanghuancoder	d1b25ed9d7	add some RecordEvent, for dygraph timeline (#30299) ... * add some RecordEvent, for dygraph timeline, test=develop * change GpuMemcpySync to memory::Copy, test=develop * fix compile problem, test=develop * fix compile problem, test=develop * fix, test=develop * fix, test=develop	4 years ago
YUNSHEN XIE	bbea5a1fa9	The new unit test cannot have the same name as the existing unit test (#29878) ... * check UT Duplicate name * fix error * Optimized log display * modified exit code	4 years ago
liym27	ff25c5b36f	Fix bug: GetAttrValue should deal with attr with attrType vector<double> (#30536)	4 years ago
WangXi	572c466d19	[Prepare for MultiProcess xpu] unified gen nccl id, refine imperative reducer (#30455)	4 years ago
ykkk2333	549855ac20	add rmsprop_op_xpu test=kunlun (#30493) ... * add rmsprop_op_xpu test=kunlun * modified rmsprop_op_xpu error code. test=kunlun	4 years ago
Zhou Wei	fb20ec9a4e	fix bug of multicard grad ncclAllReduce (#30553)	4 years ago
Zhen Wang	f30d00553a	Fix the compiling error of update_loss_scaling when using cuda9. (#30538)	4 years ago
Leo Chen	81217a94d8	unify calling cudaSetDevice (#30470) ... * unify calling cudaSetDevice * fix compile	4 years ago
pangyoki	00554b3f6b	fix error message of Inplace strategy (#30520)	4 years ago
Leo Chen	7043b8cfc6	support layer_norm fp16 in dygraph amp (#30430) ... * support layer_norm fp16 in dygraph amp * add ut * refine code	4 years ago
wanghuancoder	59ad6ff3e3	delete empty line of pybing.cc, test=develop (#30529)	4 years ago
hutuxian	e207fe6385	Ascend Framework Part2: pybind files (#30410)	4 years ago
hutuxian	40ede12631	Ascend Framework Part1: OP & Wrapper (#30281)	4 years ago
liuyuhui	843dc3cdbd	[Kunlun]PR3: add xpu executor, multi xpu card train function optimization (#30317)	4 years ago
QingshuChen	8489d4f76f	optimize batch_norm & pool op for kunlun (#30490)	4 years ago
wanghuancoder	bd97192274	if pybind.cc changed, generate total report, test=develop (#30514)	4 years ago
taixiurong	5e5c2827a3	fix range op crash in dygraph xpu place (#30469)	4 years ago
JZ-LIANG	16ba0abc79	Recompute Offload: fixed bug in memcpy (#30484)	4 years ago
guofei	11e78ebaa3	Modify the calculation logic of LambOptimizer (#29313) ... * Modify the calculation logic of LambOptimizer	4 years ago
Adam Osewski	c5ffad126c	[oneDNN] Refactor fuse pass helper functions to one place. (#30460) ... * Move pass tester helper functions to single common place. * Use helper functions in two more fuse pass tests.	4 years ago
Zhang Ting	c9a334e1b3	add VecCastCUDAKernel (#30296)	4 years ago
pangyoki	13d757362c	Add Inplace strategy (Output reuse Input Varbase) in dygraph (#30103) ... * add view strategy on squeeze,unsqueeze,reshape,flatten * add squeeze unittest * add unittests * use View strategy as name rather than Reuse Allacation * fix view api doc * fix format * use core.ops when input of reshape2 is Tensor * fix test_cross_entropy_loss error because of reshape2 * fix test_cross_entropy_loss error because of reshape2 * add inplace strategy * add elementwise_add sub * let backward op not use inplace * grad op do not use inplace * fix memory increase error and add leaf error message * delete selected_rows * change op_function * little change * solve HandleViewBetweenInputAndOutput * add unittest and leaf error message * merge view error * optimize op_function_generator format and support sum inplace op * fix format of basic_engine * fix format for framework * little change of variable wrapper * add reshape, squeeze, unsqueeze, scatter api * add relu elu tanh softmax inplace api * fix test_squeeze_op unittest * fix test_relu_op unittest * fix comment problems * delete sample code of inplace api * add reference of grad_pending_nodes in basic_engine * fix unittest name * add inplace apis into wlist * fix error message * add PADDLE_ENFORCE for set grad op twice * fix head file error	4 years ago
Yang Zhang	008b0a8b56	Fix float64 bug in layer norm (#30452) ... built-in `rsqrt` is shadowed	4 years ago
石晓伟	715d862868	export global google flags to users, test=develop (#30448)	4 years ago
Wojciech Uss	88fc7a7d68	fix cache key for inplaced elementwise ops (#30404)	4 years ago