ceci3
|
0af00a0541
|
test=develop
|
6 years ago |
ceci3
|
d3656ff304
|
test=develop
test=develop
|
6 years ago |
ceci3
|
5f343b0e3a
|
test=develop
|
6 years ago |
ceci3
|
415d74a08e
|
Merge branch 'develop' of https://github.com/PaddlePaddle/Paddle into npair_loss0
|
6 years ago |
ceci3
|
a80555a3a5
|
test=develop, change import
|
6 years ago |
Tao Luo
|
2b63317393
|
Merge pull request #15601 from luotao1/transformer_ut
add analyzer_transformer_test
|
6 years ago |
ceci3
|
d35367d296
|
Merge branch 'develop' of https://github.com/PaddlePaddle/Paddle into npair_loss0
|
6 years ago |
ceci3
|
60bfcb8b30
|
test=develop, change import
|
6 years ago |
Xin Pan
|
81773d0b1c
|
add imperative and declarative mode testbase and example
test=develop
|
6 years ago |
Zhen Wang
|
7ea5990ca6
|
update some details. test=develop
|
6 years ago |
Zhen Wang
|
bf807d69a4
|
avoid ce fails on windows.
|
6 years ago |
luotao1
|
aeee4cbe71
|
add compare between zerocopy and analysis
|
6 years ago |
Zhen Wang
|
6e0311ee8c
|
Merge pull request #16084 from wzzju/fix-quantization-pass-ut
Avoid ce fails on windows.
|
6 years ago |
Liu Yiqun
|
6bb84b74b2
|
Change the download and compress command of cmake.
test=develop
|
6 years ago |
ceci3
|
3e0eb55515
|
Merge branch 'develop' of https://github.com/PaddlePaddle/Paddle into npair_loss0
test = develop, update API.spec
|
6 years ago |
tensor-tang
|
14a764c930
|
simplify the jitkernel templates and tests
test=develop
|
6 years ago |
Jiabin Yang
|
e233b91a23
|
test=develop, fix using create_parameter with attr set to False error (#16115)
|
6 years ago |
ceci3
|
8b86c12e46
|
test=develop, update API.spec
|
6 years ago |
Tao Luo
|
25ca2ca001
|
change init_idx to INT32 in transformer_test
test=develop
|
6 years ago |
ceci3
|
c8610739c3
|
Merge branch 'develop' of https://github.com/PaddlePaddle/Paddle into npair_loss0
|
6 years ago |
ceci3
|
23a9035b21
|
test=develop, update doc
|
6 years ago |
Yiqun Liu
|
66ead07ef9
|
Make parent_idx a dispensable output for beam_search op to support models saved by older paddle version. (#16106)
test=develop
|
6 years ago |
Jiabin Yang
|
0217058321
|
test=develop, fix layers bug (#16099)
|
6 years ago |
Yihua Xu
|
0a45441a84
|
Fix the node's order issue when the content of graph is changed (#16088)
* Fix the node's sort issue when the graph is changed.
test=develop
* Clean code
test=develop
|
6 years ago |
nhzlx
|
2891070c66
|
cant not pass ci
add if use static engine for trt
test=develop
|
6 years ago |
lidanqing
|
717bbc087b
|
Add INT32 support. INT32 in last switch case
test=develop
|
6 years ago |
nhzlx
|
bcd7b9931d
|
fix wget error
test=develop
|
6 years ago |
nhzlx
|
93edcd773b
|
7 refine zero copy
update trt in docker file
test=develop
|
6 years ago |
nhzlx
|
4b59646ed1
|
fix comments and fix cpplint
test=develop
|
6 years ago |
nhzlx
|
5863c86143
|
6. delete useless predictor id
test=develop
|
6 years ago |
nhzlx
|
f3d164faad
|
5. add static trt load model
1). add static trt load model
2). fix bug: when device_id is not 0, the trt will have a bug
test=develop
|
6 years ago |
nhzlx
|
31008100ba
|
4. do the trt_engine optim during init.
add simple static mode loading
test=develop
|
6 years ago |
nhzlx
|
4f77248dd8
|
3. when runing in trt mode, do not allocate memory for parameters in fluid.
test=develop
|
6 years ago |
nhzlx
|
8c17190279
|
2. TRTEngine using stream only when execute.
|
6 years ago |
nhzlx
|
88c24baa25
|
add static model load for trt
1. bind trt input and output to fluid tensors
|
6 years ago |
Tao Luo
|
e5e7e9b865
|
Merge branch 'develop' into transformer_ut
|
6 years ago |
Yiqun Liu
|
5bde120243
|
Make parent_idx a dispensable output for beam_search op to support models saved by older paddle version. (#16106)
test=develop
|
6 years ago |
Tao Luo
|
6f2581e4c5
|
Merge pull request #16090 from lidanqing-intel/paddle-int32
Add PaddleDType INT32 support
|
6 years ago |
Jiabin Yang
|
4f3c8a41be
|
test=develop, fix layers bug (#16099)
|
6 years ago |
Yihua Xu
|
40f1dd818b
|
Fix the node's order issue when the content of graph is changed (#16088)
* Fix the node's sort issue when the graph is changed.
test=develop
* Clean code
test=develop
|
6 years ago |
Zhaolong Xing
|
3d63aa0a11
|
Merge pull request #15729 from NHZlX/add_static_model_load_for_trt
Four points for enhancing Paddle-TRT
|
6 years ago |
Zhen Wang
|
3cf0ee414d
|
update some details. test=develop
|
6 years ago |
jerrywgz
|
b0e3c02410
|
Merge pull request #15952 from jerrywgz/fpn_ops
add distribute fpn proposals op, test=develop
|
6 years ago |
ceci3
|
e1c28e7f3c
|
Merge branch 'develop' of https://github.com/PaddlePaddle/Paddle into npair_loss0
test=develop
|
6 years ago |
tensor-tang
|
802f362ac4
|
unify the kernelfuncs cache and add unit test
test=develop
|
6 years ago |
nhzlx
|
a9ed427749
|
cant not pass ci
add if use static engine for trt
test=develop
|
6 years ago |
sneaxiy
|
d0f8d94ca4
|
try to fix unittest
test=develop
|
6 years ago |
Yiqun Liu
|
f31d515ce3
|
Enhance the op benchmark: (#16066)
- Support setting attr in config
- Support setting dtype and initializer for input in config
test=develop
|
6 years ago |
tensor-tang
|
2e7fea2b7f
|
polish the cast op doc (#16078)
* polish the cast op doc
test=develop
* follow comments
test=develop
* fix api.spec
test=develop
|
6 years ago |
luotao1
|
fad06cb928
|
unify ZeroCopy in analysis_test
|
6 years ago |