mindspore-ci-bot
|
b619af917f
|
!7166 gpu no broad cast kernel dims exceed
Merge pull request !7166 from chenweifeng/broadcast-grad-dims-exceed
|
5 years ago |
mindspore-ci-bot
|
483f1aca9d
|
!7106 GPU change kernel shape to size_t
Merge pull request !7106 from VectorSL/gpu-size_t
|
5 years ago |
wilfChen
|
3eae63e4e9
|
gpu no broadcast kernel dim exceed
|
5 years ago |
VectorSL
|
f6d5508942
|
addn use intermediate results
|
5 years ago |
VectorSL
|
447a45dbe7
|
change gpu kernel shape to size_t
|
5 years ago |
VectorSL
|
f36c2721af
|
gpu add combine cast fusion
|
5 years ago |
VectorSL
|
ad0a69a60e
|
sqrt rsqrt add fp16
|
5 years ago |
mindspore-ci-bot
|
d60033c8db
|
!6381 Add dtype float16 that erf and erfc should support
Merge pull request !6381 from peixu_ren/custom_gpu
|
5 years ago |
mindspore-ci-bot
|
f6f7815fa2
|
!6440 add sin cos gpu-op
Merge pull request !6440 from wukesong/gpu-sin-cos
|
5 years ago |
wukesong
|
f9a865fd42
|
add GPU operator
|
5 years ago |
mindspore-ci-bot
|
076d8ae530
|
!6458 GPU codex fix
Merge pull request !6458 from VectorSL/codex-fix
|
5 years ago |
mindspore-ci-bot
|
1dc71651ae
|
!6437 clear warning
Merge pull request !6437 from baihuawei/clear-warnings
|
5 years ago |
VectorSL
|
9e6bd72e04
|
fix codex
|
5 years ago |
baihuawei
|
e0c063704c
|
clear warnings
|
5 years ago |
peixu_ren
|
8132e56417
|
Add dtype float16 that erf and erfc should support
|
5 years ago |
mindspore-ci-bot
|
3f0ac45954
|
!6206 new add gpu ops sqrt_grad and rsqrt_grad.
Merge pull request !6206 from linqingke/new_ops
|
5 years ago |
linqingke
|
dda3176fca
|
new add sqrt_grad and rsqrt_grad.
|
5 years ago |
mindspore-ci-bot
|
c9fa006b92
|
!6308 [MS][GPU][CUDA] NMS_Pass Kernel performance improvement
Merge pull request !6308 from danishnxt/GPU_three
|
5 years ago |
peixu_ren
|
fdd2d8209f
|
Support erf and erfc ant GPU backend
|
5 years ago |
Danish Farid
|
8c7cc7943d
|
NMS perf boost
fix:
revert the hash to orig
comment fix
remove area array
changed first workspace_size variable initialization
fixed variable name cases
|
5 years ago |
Peilin Wang
|
f020e19636
|
add int32 support to greater gpu kernel
fix ci
|
5 years ago |
mindspore-ci-bot
|
a3d0ddb4db
|
!5779 tenoradd profiling
Merge pull request !5779 from chenweifeng/broadcast-refactor
|
5 years ago |
wilfChen
|
6ebe132cd3
|
broadcast refactor
|
5 years ago |
mindspore-ci-bot
|
b9345d1d34
|
!5775 fix categorical in GraphMode
Merge pull request !5775 from baihuawei/0902
|
5 years ago |
baihuawei
|
92f1855a79
|
fix categorical in GraphMode
|
5 years ago |
limingqi107
|
5058e844cd
|
gpu inceptionv3 optimize
|
5 years ago |
mindspore-ci-bot
|
749979e7c4
|
!5458 NMS GPU OP Performance improvement
Merge pull request !5458 from danishnxt/GPU_two
|
5 years ago |
limingqi107
|
109e2e9bcc
|
modify the format info of tensorAdd
|
5 years ago |
danish
|
7d7fa760a0
|
reduce based nms final pass - speed improv
refactored faster nms
refactored faster nms + typo fix
added box flipping choice
set choice to true for testing - yz
switching back
new test file
|
5 years ago |
limingqi107
|
ff6b64a598
|
gpu GoogleNet performance optimize
|
5 years ago |
mindspore-ci-bot
|
9297ba0a8d
|
!5048 fix gpu multinomial
Merge pull request !5048 from baihuawei/0821
|
5 years ago |
baihuawei
|
c085c5f071
|
add multinomial
|
5 years ago |
mindspore-ci-bot
|
8e360888d0
|
!4590 fix gpu matmul fp32 accuracy
Merge pull request !4590 from qujianwei/master
|
5 years ago |
baihuawei
|
772e14d00d
|
add categorical
|
5 years ago |
mindspore-ci-bot
|
2b4febb430
|
!4436 Refactor uniform ops in GPU context
Merge pull request !4436 from peixu_ren/custom_gpu
|
5 years ago |
peixu_ren
|
5dd4933328
|
Refactor uniform ops in GPU context
|
5 years ago |
danish
|
97f08e74ec
|
nms_sorting fix
lint py fix 2
nms_py_file test value fix
lint fix
|
5 years ago |
qujianwei
|
c21ffc0317
|
fix gpu matmul fp32 accuracy
|
5 years ago |
Peilin Wang
|
571094f473
|
added type support for transpose and maxgrad
fix pylint
addressed code review comment
|
5 years ago |
mindspore-ci-bot
|
1856fb6af1
|
!3800 add gpu multinomial backend
Merge pull request !3800 from baihuawei/multinomial-c
|
5 years ago |
wilfChen
|
89ce0bdb78
|
maximumgrad
|
5 years ago |
baihuawei
|
40748a30c7
|
add multinomial backend
|
5 years ago |
ZPaC
|
1dcc34e785
|
Add GPU div kernel
|
5 years ago |
mindspore-ci-bot
|
eb84ae4593
|
!4048 Fix broadcast, scatternd, reduce ops.
Merge pull request !4048 from linqingke/new_ops
|
5 years ago |
mindspore-ci-bot
|
4554a80807
|
!4074 fix cumsum bug
Merge pull request !4074 from baihuawei/0806
|
5 years ago |
baihuawei
|
6053b85807
|
fix cumsum
|
5 years ago |
linqingke
|
fb405ee6f4
|
broadcast, slice, scatter_nd ops optimizer.
|
5 years ago |
mindspore-ci-bot
|
fea930f7aa
|
!4088 make gpu op Less to support int32
Merge pull request !4088 from TuDouNi/master
|
5 years ago |
root
|
3b41023a6b
|
add int32 cal for less gpu
|
5 years ago |
mamba_ni
|
96642a76fd
|
support cusolver AND OPS cholesky_solve
fix bug
clang-format
format fix
|
5 years ago |