Commit Graph

30192 Commits (73a6fa3ed0fe2bbbfe72c05f42faabccd3bbadb7)
 

Author SHA1 Message Date
Jacek Czaja 25fc2a1fdb
[oneDNN] Added Elementwise Mul grad fp32/bf16 (#31647)
4 years ago
Chen Weihang 878e117b6d
[CustomOp] Support float16 in custom op (#31725)
4 years ago
ronnywang c9e1d9dc31
[ROCM] fix test_rnn_op (#31735)
4 years ago
zlsh80826 1c67cf0c98
run radix sort of proposals layer on context stream (#31631)
4 years ago
Chen Weihang e429deb0c4
[CustomOp] Support attribute in infershape function (#31713)
4 years ago
Adam Osewski a4a2b77def
[oneDNN] lookup_table op with support for BF16 data type. (#31558)
4 years ago
zlsh80826 c86e771e94
NMS Performance Optimization (#31634)
4 years ago
zlsh80826 50cafa0b0c
remove redundant sync, set collect/dist kernel to context stream, sub_lod memcpy opt (#31641)
4 years ago
cc 1d197f6c97
[dgraph qat] Refine calculating output scale of dygraph qat (#31710)
4 years ago
ronnywang 420527f0d9
[ROCM] fix layer_norm, norm, p_norm, test_sequence_softmax_op, test_math_op_patch_var_base (#31709)
4 years ago
Chen Weihang 87852616aa
[CustomOp] Support complex dtype in custom op (#31657)
4 years ago
zlsh80826 fe241fd02f
[Paddle-TRT] gather converter (#31640)
4 years ago
zlsh80826 4ea3427865
[Paddle-TRT] support batch axis concatenation when using dynamic shape (#31627)
4 years ago
Zhou Wei d4282ea97e
fix multi cuda environment bug (#31694)
4 years ago
Chengmo 09482ddec4
【Paddle.Fleet】Fix one ps gradient clip (#31664)
4 years ago
Kaipeng Deng 740359edaf
remove useless import (#31700)
4 years ago
Zhang Ting 7f50bb7ec1
support NHWC for temporal_shift op (#31642)
4 years ago
liym27 402288ad65
In __getitem__, convert integers to int64 Tensor not int32 to be compatible with Lite(#31658)
4 years ago
Chen Weihang 2fbe9b097a
[CustomOp] Remove Eigen dependencies of float16 (#31669)
4 years ago
cc 19592d2b71
Refine dygraph qat, test=develop (#31680)
4 years ago
Zhou Wei 4c0c55bba1
support Geforce RTX 30+ GPU (#31529)
4 years ago
YUNSHEN XIE cdc5a55ac1
turn off added ut check on windows (#31660)
4 years ago
Qi Li d9b50f664f
[ROCM] update ci scripts and dockefile, test=develop (#31551)
4 years ago
YUNSHEN XIE 1a6e3b04cd
Second optimization of retry method (#31646)
4 years ago
wuhuanzhou 41e9ecfd1f
Optimize compilation with Ninja (#31449)
4 years ago
yiak c1b1ccfbf5
Update tinyformat.h (#31612)
4 years ago
gongweibao 9c624b16d5
Extend unittest time of (#31570)
4 years ago
YUNSHEN XIE 580442ceba
fix wget with no proxy on windows (#31505)
4 years ago
ronnywang da10c5cf8b
[ROCM] fix softmax_with_cross_entropy_op, test=develop (#31629)
4 years ago
LielinJiang 75433126df
Fix summary bug when calaculating output shape (#31549)
4 years ago
ShenLiang c3634c6b0a
fix amp bug of fleet (#31532)
4 years ago
Chen Weihang 027b574a0e
[CustomOp] Remove the dependence of the underlying data types on eigen (#31602)
4 years ago
WangXi 9066b74f58
c_gen_nccl_id add SocketServer to persit server (#31589)
4 years ago
Kaipeng Deng a32e8bf1e7
DataLoader supprot dict str (#31481)
4 years ago
Chen Weihang 30a627aaf3
Normalized function parameter writing (#31588)
4 years ago
Pei Yang cac9635a67
[Paddle-TRT] Fix engine key in trt int8 calibration (#31513)
4 years ago
Shang Zhizhou 50ac7dbfd0
Trt elementwise plugin serialize (#31587)
4 years ago
guofei ef0dd3efed
Support loading parameters from checkpoint to save quantized model (#31419)
4 years ago
whs da9dda5c9b
Make CreateProgramDesc more robust (#31543)
4 years ago
hong 99dcd66508
try to fix imperative orc unitest error; test=develop (#31568)
4 years ago
Qi Li 3d5aa9d10a
[ROCM] fix conv2d and conv3d op, test=develop (#31553)
4 years ago
YUNSHEN XIE f302bb4f8b
help timeout ut debug (#31500)
4 years ago
Chen Weihang 95cceb2dd7
[CustomOp] Support duplicable op input and output (#31535)
4 years ago
Aurelius84 def27bc801
[Dy2stat]Fix bug with static_convert_var_shape in locals scope (#31556)
4 years ago
YUNSHEN XIE 49c3d2a97b
modified show_ut_retry_result (#31528)
4 years ago
LielinJiang ac493f2c72
Update comments for API `RandomResizedCrop` (#31539)
4 years ago
lidanqing 0f1e7e3d52
[Bug fix] Different machine generate different binary file, remove md5 check (#31482)
4 years ago
jiangcheng 9ed6c895f1
optimize range op by place parameters on cpu rather than gpu, test=develop (#30811)
4 years ago
Thunderbrook 3789a69923
solve bug in heter mode (#31531)
4 years ago
chajchaj 6148b87f9d
add softmax_switch for softmax_with_cross_entropy_op, test=develop (#31428)
4 years ago