Commit Graph

4631 Commits (8ab29f4beae4e216f3043ce8420b063028355c31)

Author SHA1 Message Date
Zhou Wei 30888ca343
Polish and Optimize the print/repr information of Layer (#29998)
4 years ago
WeiXin f3a2392662
Extend the timeout for the (#30151)
4 years ago
Zhou Wei 9c99d37906
fix unittest failed on windows (#29837)
4 years ago
liym27 9922bd4125
Fix bug: In dynamic mode, if start or end is negetive, __getitem__ return wrong result(#30003)
4 years ago
ceci3 334247791a
add attribute for batch_norm (#29950)
4 years ago
Jiaqi Liu 2e8425b693
Fix beam search bug (#29824)
4 years ago
WeiXin f43e1d8c57
Support storage of large parameters (#29988)
4 years ago
chentianyu03 666e665132
change the kron gradient when complex types (#29995)
4 years ago
WangXi ab04997846
[fleet] combine amp and gradient merge, test=develop (#30086)
4 years ago
WangXi ee16006b5d
Optimization grad merge performance (#29784)
4 years ago
lilong12 9e51e3833f
update, test=develop (#30047)
5 years ago
chentianyu03 e012930aa3
complex gradient matmul (#29966)
5 years ago
lilong12 b0bd93de00
Disable gloo by default (#29805)
5 years ago
ShenLiang b6fd262951
fix gather nd for untest (#30037)
5 years ago
lilong12 2bc5121da8
add the paddle.distributed.split api (#29970)
5 years ago
cc c3c064a8fc
Add mkldnn nearest_interp and bilinear_interp op (#30016)
5 years ago
tangwei12 ed856d254e
fix ut (#29989)
5 years ago
Chen Weihang a1d9a14e89
support grad accumulated across batch (#29942)
5 years ago
liuyuhui bb20dcfc1a
[Kunlun] bug fix of PR2: Support MultiDevicePass and BKCL in parallel executor (#29961)
5 years ago
XiaoguangHu 726c78f293
clean redundant API alias in 2.0 - part 1 (#29928)
5 years ago
liym27 14bd77f941
[Windows CI test] Enable unittest test_optimizer_in_control_flow and remove unnecessay code (#29851)
5 years ago
littletomatodonkey 5c162fe66e
fix reg api ut fail (#29921)
5 years ago
Leo Chen a4b9daf97c
fix optimizer dtype (#29917)
5 years ago
liuyuhui 4427df37cf
[Kunlun] PR2: Support MultiDevicePass and BKCL in parallel executor (#29574)
5 years ago
LielinJiang 0b74428db8
Fix Conv2DTanspose bug when padding='same' (#29915)
5 years ago
lilong12 01950ceb42
fix the bug in pipeline data parallelism (#29731)
5 years ago
Chen Weihang a6072055be
[Complex] Handle complex to real after type promotion (#29855)
5 years ago
Chen Weihang 1a304e6c06
[Complex] Add support for complex grad accumulated (#29889)
5 years ago
guofei 80eb77788f
Skip Windows Multi-GPU test of test_fetch_lod_tensor_array (#29508)
5 years ago
Leo Chen 6b258317cb
fix TransferInplaceBack (#29830)
5 years ago
QingshuChen 59b47f3b32
feat: support check_nan_inf for kunlun/xpu device (#29694)
5 years ago
wawltor 7498df2587
add the cumsum unit test for the develop (#29881)
5 years ago
wanghuancoder 26f9ab70f7
if PR have no .py files, do not use 'python coverage run', to speedup unit test (#29739)
5 years ago
Tao Luo 5d130d5670
Revert "fix conv2d int8 windows UT (#29528)" (#29869)
5 years ago
tangwei12 032414ca2a
[Feature] one ps (3/4) (#29604)
5 years ago
jakpiase edc06c6a1b
Added fc + activation fuse pass (currently only gelu, sigmoid and tanh are supported) (#29772)
5 years ago
Chen Weihang 0e0bb1b97d
replace exit method (#29862)
5 years ago
lidanqing 067d7f1d0d
fix conv2d int8 windows UT (#29528)
5 years ago
liym27 97e75ad0f5
[setitem] Support Tensor setitem in static mode (#29708)
5 years ago
YUNSHEN XIE 24ce051a84
remove duplicate ut reload (#29810)
5 years ago
ceci3 c4eb5d0378
fix unittest timeout (#29820)
5 years ago
chentianyu03 ddfc3d2c2f
change grad elementwise_mul for complex types (#29757)
5 years ago
chentianyu03 2a260d9b0e
change the grad of div when complex types (#29804)
5 years ago
Guo Sheng 356efd36fa
Remove test_rnn_decode_api from disable list. (#29814)
5 years ago
TTerror 82aa01c373
add nearest_interp_v2 on kunlun (#29725)
5 years ago
whs 82630408b4
Support double backward rsqrt (#29589)
5 years ago
xiaoting 55725cd2e1
fix for timeout, test=develop (#29788)
5 years ago
LielinJiang a94c3cbbf3
register cudnn conv double grad for depthwise conv (#29807)
5 years ago
liym27 0cc42e34c6
Migrate 4 APIs about array to paddle.tensor.* (#29565)
5 years ago
liym27 41a7b07159
[Dy2Stat] Fix bug for loop: a variable is used and created in loop, but used before created (#29769)
5 years ago
LielinJiang e5af650b71
Add double grad for conv_transpose (#29706)
5 years ago
Wojciech Uss 6ef8129dcc
upgrade oneDNN with GRU INT8 optimizations (#28420)
5 years ago
Huihuang Zheng dfffee8a5d
[Dy2stat] Enable jit.save to Save Without Running (#29579)
5 years ago
liym27 a0b60716f1
[Dy2Stat] Support grammar: for ele in var[idx] (#29541)
5 years ago
chentianyu03 b59b6d7ae6
Complex op test (#29753)
5 years ago
liym27 096c048b45
Fix unitest test_slice (#29740)
5 years ago
Huihuang Zheng 2e788bd81e
Reduce batch size ot fix CPU memory, test=develop (#29736)
5 years ago
chentianyu03 71063b8137
add conj op for complex types (#29527)
5 years ago
Chen Weihang 6cfa59de1b
[Complex] Add real & imag op and api for complex tensor (#29672)
5 years ago
TTerror af8ded773a
update activation op on kunlun (#29577)
5 years ago
ceci3 cc387159f3
add pad and concat double grad (#29549)
5 years ago
liuyuhui f13c3a9cd7
[Kunlun] PR1:Support one Kunlun card training in parallel executor (#29337)
5 years ago
YUNSHEN XIE d0b789d27f
disable ut test_cumsum_op (#29613)
5 years ago
Jack Zhou 84bae27779
fix wmt14 doc, remove backward, add bidirect direction in rnn api (#29633)
5 years ago
YUNSHEN XIE 2926e74326
New UT should not exceed 15s (#29492)
5 years ago
Chen Weihang f02aece1f0
Add complex dtype op (add) test example (#29603)
5 years ago
AshburnLee efea540ca9
Add tf32 support for A100 tensor core acceleration for cuBLAS (#28732)
5 years ago
lijianshe02 7779768b53
add transpose double grad test=develop (#29600)
5 years ago
ShenLiang 1efef8baed
Fix bug of matmul_v2 for broadcast case (#29599)
5 years ago
qingqing01 8d549fc85d
Add clip double grad (#29590)
5 years ago
Tao Luo 81acc3278c
disable test_parallel_executor_profiler in cuda 10.1 (#29581)
5 years ago
wangchaochaohu ac4bae8ee9
elementwise_add_grad Op optimization (#29575)
5 years ago
WangXi 467c716963
gen nccl id use socket (#29431)
5 years ago
Bai Yifan d72604cd46
fix unittst unstable issue on ci machine (#29588)
5 years ago
QingshuChen 79a41a9ed6
support roi_align & affine_channel for kunlun (#29561)
5 years ago
liym27 0cad1152f4
[Dy2Stat] 1. Fix bug of for-range stmts. 2. Support that step value is negative in for-range stmts (#29519)
5 years ago
Huihuang Zheng 831e9135b9
Fix Windows Unittest (#29543)
5 years ago
GeminiCarrie 08f24a3108
Fix precision problem (#29567)
5 years ago
Chen Weihang c1a26e2a05
fix train eval set error in static mode (#29540)
5 years ago
taixiurong 760d015c14
add xpu ops for training transformer in kunlun (#29539)
5 years ago
Leo Chen 0fdd365665
Add fast path for dropout when p == 0 (#29553)
5 years ago
lijianshe02 bd29052e33
fix random seed in nll_loss unitest test=develop (#29538)
5 years ago
joanna.wozna.intel 0ce6d7fa77
Fix bf16 activations test for softmax and gelu (#29502)
5 years ago
huangxu96 4001979309
Add ReserveSpace in dygraph batch_norm. (#29221)
5 years ago
arlesniak b781953ef5
[oneDNN] Fix flags use test for #29080, assert condition more general (#29493)
5 years ago
Zhen Wang 5ac71b36fb
Remove tensor copy in the update_loss_scaling op. (#29426)
5 years ago
Zhou Wei e74e1a226c
support deepcopy for Layer/Tensor/Paramerbase (#29387)
5 years ago
joejiong 50d3117d30
Add random_split and Subset dataset (#29291)
5 years ago
joejiong 87e75a77c2
Add tangent operator (#29207)
5 years ago
Wei Shengyu dc8bb76c68
remove addcmul (#28937)
5 years ago
Zhong Hui f459dd9634
fix abs double grad unittest (#29478)
5 years ago
ShenLiang 2ef9e0e23c
Rebuild group automatically in dynamic graph distributed (#29255)
5 years ago
procr 3a0558339d
support mobilenet for kunlun (#29458)
5 years ago
ShenLiang 311b3b44fc
Fix the bug where embedding can‘t be processed correctly in reducer (#29485)
5 years ago
Pei Yang 2480bdef6c
change hard_swish from plugin to layer (#29177)
5 years ago
lilong12 b122d0bb76
Fix bug in gloo that gloo initialization hangs (#29447)
5 years ago
taixiurong ecca6585cd
1. fix elementwise ops'bug 2. fix softmax_with_cross_entropy_op 3. add biliner_interp_op (#29448)
5 years ago
LoveAn 03b42d9fa7
fix unittest on windows, test=develop (#29365)
5 years ago
ShenLiang 22e6b9e373
Fix the ut of matmulv2 for broadcast case (#29461)
5 years ago
TTerror a5fcc4b545
update reduce_sum op on xpu (#29367)
5 years ago
Zhang Ting 6296f4ed09
revert cast eigen kernel (#29427)
5 years ago
Leo Chen a040c055a5
fix layer_norm accuracy (#29434)
5 years ago
Shang Zhizhou 225a9c4ed8
Fix unittest (#29412)
5 years ago
Pei Yang f860de4af7
support clip op trt converter (#29411)
5 years ago
Bai Yifan 87bb726258
Add deform_conv2d,DeformConv2D (#29364)
5 years ago
chentianyu03 64e4e17f0c
remove complexvariable (#29390)
5 years ago
chajchaj 79e6086743
change shape of output in cross_entropy, test=develop (#29220)
5 years ago
chentianyu03 879e913b6d
Make transpose, trace, kron, reshape, sum op support complex type (#29321)
5 years ago
liym27 5f84d0b375
Fix bug: delete wrong check_type of paddle.concat and support LoDTensorArray (#29306)
5 years ago
卖鱼的哲学 074065e5de
fix expand/uniform_random && concat/transpose to new api on xpu (#29280)
5 years ago
gongweibao 8989053443
Fix bug of test_fleet_launch_async.sh (#29332)
5 years ago
Huihuang Zheng 8f7627907c
[Dy2stat] Reduce Exception Type for Better Error Message (#29268)
5 years ago
liym27 61a8f2874f
[Dy2Stat] Fix bug: Do not use gast.Subscript to replace gast.Name in when transforming for_enumerate_loop (#29310)
5 years ago
liym27 b10ecd9d3a
[inplace] Add ShareHolderWith for class Variable and SharePlaceholderWith in VarBase.detach() to share the same Tensor/SelectedRows (#29267)
5 years ago
Chen Weihang 9ad800ebb2
Support type promote for basic math ops (quantum required) (#29265)
5 years ago
YUNSHEN XIE 28164b266f
disable test_rnn_decode_api and test_complex_matmul on windows (#29252)
5 years ago
Leo Chen b58cfff89d
use has_grad instead of train_mode (#29309)
5 years ago
Aurelius84 67c700b479
[Dy2Stat] Add cache for Executor and Context in run_program_op (#28421)
5 years ago
ShenLiang d6753e1e6d
fix matmulv2 for windows (#29327)
5 years ago
liym27 b9a8ebd50f
[Dy2stat] Add a decorator paddle.jit.not_to_static to support that not to convert a function in Dynamic-to-Static. (#29253)
5 years ago
ShenLiang c00af94435
fix matmulv2 for windows (#29302)
5 years ago
LielinJiang b9f1f4343b
Move temporal_shift to paddle.nn.functional (#29261)
5 years ago
Chen Weihang a2e9d95a4a
change test_imperative_signal_handler_to_exclusive (#29283)
5 years ago
Zhen Wang be3777a50a
Add pure fp16 training with master weights. (#27712)
5 years ago
chentianyu03 976961de6d
fix random failed of complex matmul (#29285)
5 years ago
furnace 7584bb5096
Layer norm fp16 (#29169)
5 years ago
Huihuang Zheng aec05d811c
[Dy2stat] Fix PaddleGan Deoldify Model Dy2stat Problems (#29226)
5 years ago
Leo Chen 116305ea4b
Improve performance of elementwise_add grad op (#29187)
5 years ago
卖鱼的哲学 07c67d5a8b
add deformable_conv op on xpu (#29234)
5 years ago
Chen Weihang 1de32f823d
Hot fix complle failed in gcc4.8 caused by complex impl (#29254)
5 years ago
ShenLiang 46b73e6cd9
Change the api of DataParallel and Fleet (#29224)
5 years ago
QingshuChen 64f29fbb70
update kunlun conv2d/softmax/elementwise implemetation (#29229)
5 years ago
Jiawei Wang b11ab12787
Fix doc (adadelta, sgd, momentum) (#29212)
5 years ago
lijianshe02 76312deb30
fix nll_loss test random fail bug test=develop (#29236)
5 years ago
Jiawei Wang a5d13d593c
Momentum Velocity init in Momentum.__init__() (#29223)
5 years ago
wanghuancoder 2b2cd1864a
revert python file coverage, delete coverage run --include, test=develop (#29230)
5 years ago
chentianyu03 8f45d14263
add complex64 and complex128 type; add +-*/@ and slice opreator for c… (#29199)
5 years ago
Zhou Wei c0a991c874
accumulate gradient for leaf tensor with previous graph and expose leaf tensor concept (#28429)
5 years ago
Chen Weihang 786e69e9c7
diable test_yolov3 in musl (#29216)
5 years ago
hong19860320 f23665e5d5
Refine the doc and unit test for Sigmoid and stanh (#29198)
5 years ago
123malin b5c6342336
Update ps gpu (#29209)
5 years ago
liym27 865a45984f
Check whether there is any inplace operation affecting gradient calculation. (#27901)
5 years ago
joejiong dc070ecfb0
Remove cast from paddle.pow api (#29134)
5 years ago
WangXi 0c2a51d240
optimizer amp, all use fp16 communication, overlap last comm and compute (#28957)
5 years ago
Chen Weihang 0b032faeee
Polish unittests details and execution conditions to adapt to MUSL (#29044)
5 years ago
Thunderbrook 4adddcc89a
add set_trainer_num api in dataset (#29133)
5 years ago
liym27 e03440812a
fix code: if y is True -> if y (#29184)
5 years ago
WeiXin 1476e1f998
save model after jit.load (#28748)
5 years ago
wanghuancoder 0239f79695
Generate code coverage reports only for incremental files (#28508)
5 years ago
Huihuang Zheng 27b4218333
[Dy2stat] Disable PaddleInference IR Optimization in test_mnist for CUDA11 (#29105)
5 years ago