Megvii Engine Team
|
bccda5c427
|
fix(mgb/imperative): fix repeat bug in trace mode
GitOrigin-RevId: 9547fc6102
|
2 years ago |
Megvii Engine Team
|
fca6c76a0e
|
fix(lite): fix input invalid bug in lar for fitting mode
GitOrigin-RevId: 45d81c9a96
|
2 years ago |
Megvii Engine Team
|
1b94380794
|
fix(dnn): fix reduce sum/mean error when b is large
GitOrigin-RevId: d1bae619b1
|
2 years ago |
Megvii Engine Team
|
c7a9909839
|
feat(cuda): add int4 ptx 256x64 mma kernel
GitOrigin-RevId: 8f7475b0f6
|
2 years ago |
Megvii Engine Team
|
cf3ca1e9a2
|
feat(cuda): add int4 ptx 128x256 mma kernel
GitOrigin-RevId: 1ae7c9f034
|
2 years ago |
Megvii Engine Team
|
1f8e930e28
|
feat(cuda): add int4 ptx 128x128 mma kernel
GitOrigin-RevId: 5a8b9c3f8e
|
2 years ago |
Megvii Engine Team
|
1a2ed8c47b
|
feat(cuda): add convbias ptx algo testcase
GitOrigin-RevId: 9ad6d4561f
|
2 years ago |
Megvii Engine Team
|
64551105f9
|
feat(cuda): add convbias ptx algo
GitOrigin-RevId: 08e9f66641
|
2 years ago |
Megvii Engine Team
|
8395a459b5
|
fix(dnn/fallback): fix naive shift multidefination error and optimize GiCvtFromInt32V4ToUint8
GitOrigin-RevId: 6660c35214
|
2 years ago |
Megvii Engine Team
|
cc21855074
|
feat(lite): load_and_run support optimize for inference
GitOrigin-RevId: d9abb8de9e
|
2 years ago |
Megvii Engine Team
|
9bbe550032
|
fix(opencl/extern_c_opr): fix cl_mem UAF issue when
run model OpenCL + ExternCOprRunner, for example
graph: part_a(OpenCL) --> part_b(ExternCOprRunner) --> part_c(OpenCL)
GitOrigin-RevId: f754b559a2
|
2 years ago |
Megvii Engine Team
|
23a3d13350
|
fix(dnn/softmax): create redcue and elemwise opr when get workspace size
GitOrigin-RevId: 476a39bdd3
|
2 years ago |
Megvii Engine Team
|
2797fcfad0
|
fix(mge/device): add missed API to __all__ scope
GitOrigin-RevId: c3ce90990f
|
2 years ago |
Megvii Engine Team
|
d7c546c90a
|
fix(mge/interpreter): regenerates tensor when its dev value is needed
GitOrigin-RevId: ed26d52ee4
|
2 years ago |
Megvii Engine Team
|
1f7bf1ada3
|
fix(opr): fix the compatilibity of elemwise multitype new mode
GitOrigin-RevId: ee58271276
|
2 years ago |
Megvii Engine Team
|
b3a7d149a0
|
feat(dnn/fallback): add some new gi api
GitOrigin-RevId: 4aede0ac6a
|
2 years ago |
Megvii Engine Team
|
198ee0686f
|
feat(mgb/trt): update tensorRT toolchain to 8
GitOrigin-RevId: d7cbb722b8
|
2 years ago |
Megvii Engine Team
|
626222c698
|
fix(test): fix test for brainpp docker env
GitOrigin-RevId: c4c2cc73d2
|
2 years ago |
Megvii Engine Team
|
fac67e7c2b
|
feat(gopt): support nchw44 global pooling with fuse_grain
GitOrigin-RevId: 4c43a149f8
|
2 years ago |
Megvii Engine Team
|
8461c8d8e7
|
fix(lite): fix ldr use lite interface error when open both fast-run and nchw44
GitOrigin-RevId: 27b29d60af
|
2 years ago |
Megvii Engine Team
|
43bd949af0
|
fix(dnn): fix cudnn include
GitOrigin-RevId: f6f9731c3e
|
2 years ago |
Megvii Engine Team
|
8abc3ab8fc
|
fix(imperative): fix convolution in rocm
GitOrigin-RevId: 9e97099fd5
|
2 years ago |
huangxinda
|
3b1101b5e9
|
feat(ci): update image
|
2 years ago |
Megvii Engine Team
|
32b31fd578
|
fix(mgb): change the check method of cuda sm code
GitOrigin-RevId: 23dbc9b574
|
2 years ago |
Megvii Engine Team
|
5f86368219
|
Revert "feat(dnn): add elemwise modes"
This reverts commit cb713ddb24 .
GitOrigin-RevId: 02adf025e6
|
2 years ago |
Megvii Engine Team
|
d2a1905ad5
|
Revert "feat(mgb): add cumprod opr"
This reverts commit 3436c3bdaa .
GitOrigin-RevId: 95ab3d1aa7
|
2 years ago |
Megvii Engine Team
|
49e14f87b5
|
feat(mgb): add cumprod opr
GitOrigin-RevId: 3436c3bdaa
|
3 years ago |
Megvii Engine Team
|
87aedc2991
|
feat(dnn): add elemwise modes
GitOrigin-RevId: cb713ddb24
|
3 years ago |
Megvii Engine Team
|
25e89d68b0
|
feat(gi/rvv): remove winograd rvv do not use FIXLEN workaround
GitOrigin-RevId: fce5103088
|
2 years ago |
Megvii Engine Team
|
fe5b1834ff
|
fix(mgb/imperative): fix the problem of occasional failure during testing of redis
GitOrigin-RevId: da1d55c70d
|
2 years ago |
Megvii Engine Team
|
b3f46734e7
|
feat(megdnn/softmax): add softmax operator in fallback
GitOrigin-RevId: 97bc32f561
|
3 years ago |
Megvii Engine Team
|
6c78e68451
|
fix(lite): fix lite memory leak
GitOrigin-RevId: 075c686162
|
2 years ago |
Megvii Engine Team
|
ff239c638a
|
feat(lite): add unit test for lar
GitOrigin-RevId: f3ba1e8a9a
|
2 years ago |
Megvii Engine Team
|
7bf1c38c74
|
fix(mgb/imperative): fix imperative code gen
GitOrigin-RevId: da9e8a280a
|
2 years ago |
Megvii Engine Team
|
c49d3070ba
|
refactor(imperative/ops): extends DnnOprCaller with template
GitOrigin-RevId: 402cba209a
|
2 years ago |
Megvii Engine Team
|
2d6476a416
|
feat(lite): add auto decide model inference format option
GitOrigin-RevId: fcbf945de5
|
3 years ago |
Megvii Engine Team
|
10a0349eca
|
feat(lite): add assert log for set_data_by_share
and set_data_by_copy. pylite network input is not
correct when input np is not continuous
GitOrigin-RevId: 1bdeae970a
|
2 years ago |
Megvii Engine Team
|
f5597d9a10
|
fix(mgb): make error infomation of input channel mismatch more readable
GitOrigin-RevId: 6f95260070
|
2 years ago |
Megvii Engine Team
|
38bd599911
|
fix(mgb): make error infomation of invalid MatMul more readable
GitOrigin-RevId: 96b922dd20
|
2 years ago |
Megvii Engine Team
|
e0d505e6bd
|
fix(mgb/dnn): fix bug that some cutlass file compile very slowly on SM86
GitOrigin-RevId: 91d7ac1927
|
2 years ago |
Megvii Engine Team
|
cc31b9db34
|
docs(mge/functional): fix vision docs
GitOrigin-RevId: f03567d8ae
|
2 years ago |
Megvii Engine Team
|
6f7649e935
|
docs(docstring): fix pad docstring
GitOrigin-RevId: 698d2f1c0d
|
2 years ago |
Megvii Engine Team
|
bf32d1e04f
|
docs(dataloader): update dataloader docstring
GitOrigin-RevId: 3e94a4bdf4
|
2 years ago |
Megvii Engine Team
|
4a32cc493a
|
docs(mge/data): update Dataset class docstring
GitOrigin-RevId: f08d818cf3
|
2 years ago |
Megvii Engine Team
|
70fc568224
|
docs(mge/data): update MNIST dataset docstring
GitOrigin-RevId: 536a46325f
|
2 years ago |
kagome1007
|
8fb062dfba
|
Merge pull request #468 from MegEngine/HuaHua404-patch-2
docs(Readme): add key features description
|
2 years ago |
kagome1007
|
e5ad3ea520
|
Merge pull request #469 from MegEngine/HuaHua404-patch-3
docs(readme): Add key features description
|
2 years ago |
Megvii Engine Team
|
58ba080d5f
|
feat(x86/rvv): make gi conv algo adapt to vv and vf model
GitOrigin-RevId: f29593be4d
|
2 years ago |
Megvii Engine Team
|
bd50e457ee
|
feat(x86/rvv): make MATRIX_MUL_GI_F32_4x12 and FP32_GEMV_MK4_GI
adapt to vv and vf model
GitOrigin-RevId: 691434c598
|
2 years ago |
Megvii Engine Team
|
5c3b4e9584
|
feat(x86/rvv): opt AlgoFP32WinogradF63_4x4_NCHW44
GitOrigin-RevId: 0cd0089982
|
2 years ago |