mengyuanli
|
af1deb1c10
|
fix bug of lite
|
4 years ago |
i-robot
|
846205cdab
|
!26096 Add int as input type for AbsGrad CPU op.
Merge pull request !26096 from hezhenhao1/master
|
4 years ago |
fangzehua
|
2429136828
|
fix stack infer and cpu op
|
4 years ago |
i-robot
|
bf39da2555
|
!26118 Fix Scalar Input Case for PyFunc
Merge pull request !26118 from jiaoy1224/arithmetic
|
4 years ago |
gongdaguo
|
136b9ba298
|
add int8 clip
|
4 years ago |
Yang Jiao
|
ef0161caa6
|
fix pyfunc scalar support
|
4 years ago |
hezhenhao1
|
be7390964c
|
Add int as input type for AbsGrad CPU op.
|
4 years ago |
i-robot
|
b4c1d771c1
|
!25855 [ms][lite][cpu] deconv optimize
Merge pull request !25855 from liuzhongkai/deconv_114
|
4 years ago |
i-robot
|
6dcf8e2048
|
!26053 [MS][LITE][develop] add fp16 kernel
Merge pull request !26053 from sunsuodong/add_fp16_kernel_b
|
4 years ago |
i-robot
|
d4c8dde6d2
|
!25923 Move IsRealKernel and IsRealCNodeKernel from AnfAlgo to AnfUtils
Merge pull request !25923 from DeshiChen/1101_anfalgo
|
4 years ago |
sunsuodong
|
a8aad920bc
|
add fp16 kernel
|
4 years ago |
i-robot
|
10a4c843ba
|
!25802 [MS][LITE] fix npu graph format error
Merge pull request !25802 from XianglongZeng/myms_new
|
4 years ago |
dayschan
|
cbb84ff580
|
Move IsRealKernel and IsRealCNodeKernel from AnfAlgo to AnfUtils
the function IsOneOfPrimitive and IsOneOfPrimitiveCNode is useful,
we can move them into anf.cc
|
4 years ago |
i-robot
|
9a3b52eda7
|
!25726 Add Pynative Test Cases for Custom AOT, Pyfunc
Merge pull request !25726 from jiaoy1224/arithmetic
|
4 years ago |
zengxianglong
|
5d8ae091ec
|
fix npu graph format error
|
4 years ago |
i-robot
|
3087ff09ec
|
!25948 mkdir for MS_COMPILER_CACHE_PATH if not exists
Merge pull request !25948 from lingyunli63/mkidr_for_ms_compiler_cache_path
|
4 years ago |
lzk
|
29dafaaa67
|
deconv optimize
|
4 years ago |
i-robot
|
770e85409e
|
!25810 [MSLITE] Fuzz test
Merge pull request !25810 from wangshaocong/fuzz_test
|
4 years ago |
i-robot
|
2b3fbd5cec
|
!25958 op debug level update
Merge pull request !25958 from liubuyu/master
|
4 years ago |
lingyunli63
|
4c3f9b2217
|
mkdir if not exist
|
4 years ago |
i-robot
|
4c46d098b9
|
!25994 Add fp64 as input type fot GPU op of ReduceMin and ReduceMean.
Merge pull request !25994 from hezhenhao1/add_double
|
4 years ago |
i-robot
|
cba3c57e20
|
!25860 [MSLITE] all gather
Merge pull request !25860 from ling/sr
|
4 years ago |
Yang Jiao
|
2b112ef656
|
add custom pynative test cases
|
4 years ago |
hezhenhao1
|
cb13783059
|
Add fp64 as input type fot GPU op of ReduceMin and ReduceMean.
|
4 years ago |
i-robot
|
a58eb682d6
|
!25778 Custom op supports no reg info
Merge pull request !25778 from looop5/reg_info_commit
|
4 years ago |
ling
|
bbdc907673
|
[MSLITE] all gather
|
4 years ago |
i-robot
|
0598bcd949
|
!25973 [MS][LITE][develop] optimize x86 init
Merge pull request !25973 from sunsuodong/x86_init
|
4 years ago |
lby
|
b7bef8d99c
|
opdebug level support [0, 1, 2, 3, 4]
|
4 years ago |
sunsuodong
|
622f672cf9
|
optimize x86 init
|
4 years ago |
i-robot
|
a3441bbfb5
|
!25863 fix dynamic shape bugs
Merge pull request !25863 from fangzehua/fix_dy_1028
|
4 years ago |
fangzehua
|
4433fbd1fa
|
fix dynamic shape
|
4 years ago |
looop5
|
b89d744e80
|
Custom op supports no reg info
|
4 years ago |
i-robot
|
3caa6cb09e
|
!25898 split generic matrix/symmetric matrix eigenvalues to eig/eigh(same with scipy)
Merge pull request !25898 from wuwenbing/master
|
4 years ago |
i-robot
|
064bb48326
|
!25924 fix json parse failed
Merge pull request !25924 from liubuyu/master
|
4 years ago |
i-robot
|
fb155743a7
|
!25921 Fix: cpu real_div support fp16.
Merge pull request !25921 from hezhenhao1/add_fp16
|
4 years ago |
i-robot
|
279a099936
|
!25138 [MS][LITE][ToD] mindir_fp16
Merge pull request !25138 from Nizzan/export_nizzan
|
4 years ago |
lby
|
2bd2a7140a
|
fix json parse failed
|
4 years ago |
hezhenhao1
|
14785097fd
|
Fix: cpu real_div support fp16.
|
4 years ago |
i-robot
|
c48778df30
|
!25683 Add GraphKernelCallback functions, and call them in AkgKernelJsonGenerator.
Merge pull request !25683 from DeshiChen/1025_genjson
|
4 years ago |
wenbean
|
45f9c7bfb7
|
use workspace memery for symmetric matrix eigenvalue/eigenvectors
split generic matrix decompress with symmetric matrix to eig and eigh ops
|
4 years ago |
i-robot
|
f7eb98e339
|
!25868 add float64 support for concat and print
Merge pull request !25868 from zhujingxuan/double
|
4 years ago |
wang_shaocong
|
9a4b7dd0f6
|
[MSLITE] Fuzz test fix
|
4 years ago |
i-robot
|
b6bf3cd641
|
!25762 Unified CPU thread pool
Merge pull request !25762 from 范吉斌/unified_thread
|
4 years ago |
zhujingxuan
|
cc1e9f1fb9
|
add float64 support
|
4 years ago |
nizzan
|
665e2a738e
|
Support mix precision mindir file
|
4 years ago |
i-robot
|
55a83cf31e
|
!25859 Support int32 as input type for Abs of GPU op, float64 as input type for isFinite GPU op.
Merge pull request !25859 from hezhenhao1/gpu_support
|
4 years ago |
i-robot
|
c7edc7d5eb
|
!25830 modify log level
Merge pull request !25830 from hwjiaorui/check-support
|
4 years ago |
i-robot
|
4771a2006e
|
!25800 Suport solve complex64/ complex128 Eigen Value and eigen vectros
Merge pull request !25800 from wuwenbing/master
|
4 years ago |
hezhenhao1
|
d61b089f6b
|
Support int32 as input type for Abs of GPU op, float64 as input type for IsFinite GPU op.
|
4 years ago |
fanjibin
|
b6acbd1055
|
unified cpu thread pool
|
4 years ago |