sneaxiy
|
e57bc4d745
|
Merge branch 'refine_elementwise_add' of https://github.com/sneaxiy/Paddle into refine_elementwise_add
|
7 years ago |
sneaxiy
|
222fbbedfb
|
Merge branch 'develop' into refine_elementwise_add
|
7 years ago |
|
sneaxiy
|
4b83afff6e
|
Merge branch 'develop' into refine_elementwise_add
|
7 years ago |
|
sneaxiy
|
b2d0ee5159
|
refine elementwise_add op
|
7 years ago |
dzhwinter
|
595a2c83ae
|
explicit gradient of elementwise_add/elementwise_sub (#11970)
* "add gradient register"
* "make some enhance"
* "better format"
* "fix typo"
* "fix reuse"
* "fix get expected kernel"
* "change the mkldnn code"
* "fix mkldnn"
* "fix mkldnn failed test"
* "add comment"
|
7 years ago |
Tomasz Patejko
|
3e876b3e49
|
MKL optimized elementwise add: fix style check
|
7 years ago |
|
Tomasz Patejko
|
9241011b31
|
MKL elementwise add backward: backward works for integral types with fall back to default impl
|
7 years ago |
|
Tomasz Patejko
|
fde47aae61
|
MKL elementwise add backward: grad inputs copied when they are not null
|
7 years ago |
|
Tomasz Patejko
|
996d12f16d
|
MKL optimized elementwise add backward: coding style fixes
|
7 years ago |
|
Tomasz Patejko
|
5a622c2906
|
MKL elementwise add backward: Initial implementation with vector copy
|
7 years ago |
|
Tomasz Patejko
|
01fb2be908
|
MKL elementwise add: default implementation used for integral types, float16 and/or GPU
|
7 years ago |
|
Tomasz Patejko
|
6f932482f4
|
MKL elementwise_add: BLAS version compiles with integral types
|
7 years ago |
|
Tomasz Patejko
|
e43c8f33cd
|
MKL elementwise add: elementwise_add uses vAdd VML function when MKL is used
|
7 years ago |
Yu Yang
|
88c22e9d1a
|
Speed up elemwise grad (#8402)
* Speed up elemwise grad
* Fix bug
* Add macro for MAX_BLOCK_DIM
|
7 years ago |
qingqing01
|
24509f4af9
|
Fix the grammar in copyright. (#8403)
|
7 years ago |
Yi Wang
|
fc374821dd
|
Correct #include path
|
7 years ago |
|
Yi Wang
|
90648f336d
|
Move file to fluid/; Edit CMakeLists.txt
|
7 years ago |
