Megvii Engine Team
|
58c8746e30
|
fix(opr): fix fast-run error in cuda
GitOrigin-RevId: 28dd187df9
|
5 years ago |
Megvii Engine Team
|
5d350fc843
|
feat(dnn/cuda): add deconv int8 and fix cutlass conv wrapper base on modify cutlass 2.4
GitOrigin-RevId: 49e0565e8a
|
5 years ago |
Megvii Engine Team
|
a3ea1f153c
|
feat(mgb/opr): add fast profile and combined Execution strategy
GitOrigin-RevId: 843dc3a790
|
5 years ago |
Megvii Engine Team
|
c82d88751a
|
fix(dnn/cuda): add cuda nchw int8 conv impl with nchw4 to fix cu111 compatibility
GitOrigin-RevId: 771968f9ac
|
5 years ago |
Megvii Engine Team
|
f2b42bf09e
|
chore(dotprod): add arm dotprod attribute for easy use
GitOrigin-RevId: 78c3e72218
|
5 years ago |
Megvii Engine Team
|
c33a717314
|
feat(dnn): repalce is_reproducible with algo attribute in opencl, cpu, rocm and cuda
GitOrigin-RevId: 86dead0a11
|
5 years ago |
Megvii Engine Team
|
9cc732f82d
|
fix(opencl): fix opencl search algo negative stride support
GitOrigin-RevId: 0642d1718d
|
5 years ago |
Megvii Engine Team
|
cd7090acbb
|
fix(opencl): enable image on mali(cl2.1)
GitOrigin-RevId: 0c670fba80
|
5 years ago |
Megvii Engine Team
|
c51a687cef
|
chore(mge): update copyright years
GitOrigin-RevId: 46104ac891
|
5 years ago |
Megvii Engine Team
|
7afa422df4
|
refactor(megdnn): refactor sub opr setter
GitOrigin-RevId: 475afb9c10
|
5 years ago |
Megvii Engine Team
|
f14e0c17e7
|
feat(mgb): add recursive for fastrun and megdnn test
GitOrigin-RevId: 743846f645
|
5 years ago |
Megvii Engine Team
|
85fa988348
|
refactor(dnn): add get_algorithm_from_desc interface
GitOrigin-RevId: 6d211ca167
|
5 years ago |
Megvii Engine Team
|
364afec033
|
chore(mge): update copyright years
GitOrigin-RevId: 3c0690bcc1
|
5 years ago |
Megvii Engine Team
|
8f7f52ae4d
|
feat(jit): add memfwd in jit executor opr
GitOrigin-RevId: b58860bbe8
|
5 years ago |
Megvii Engine Team
|
dfb2b2ce49
|
fix(dnn): change pooling window size smaller than padding constraint to log_error
GitOrigin-RevId: c3cda68f6d
|
5 years ago |
Megvii Engine Team
|
a85531dd0f
|
feat(mgb/opr): add tqt opr
GitOrigin-RevId: 49c62cd532
|
5 years ago |
Megvii Engine Team
|
61f917fb8e
|
feat(dnn/cuda): add impl for fusing warp perspective and dimshuffle
GitOrigin-RevId: 51e025973f
|
5 years ago |
Megvii Engine Team
|
eb826422c4
|
fix(dnn): forbid pooling window size smaller than padding
GitOrigin-RevId: 9ad61c409d
|
5 years ago |
Megvii Engine Team
|
fc0fcd2f7f
|
chore(winograd): remove winograd transform code
GitOrigin-RevId: 78c3cfceae
|
5 years ago |
Megvii Engine Team
|
d1adc9a22f
|
fix(dnn): fix opencl algo search
GitOrigin-RevId: 25997d0ef1
|
5 years ago |
Megvii Engine Team
|
3bf73ff16f
|
feat(dnn): add cuda preprocess fusion
GitOrigin-RevId: d789c99e59
|
5 years ago |
Megvii Engine Team
|
86cf7490ec
|
feat(dnn/aarch64): add quantizeds4 matmul int4x4x16_k8x8x8
GitOrigin-RevId: 7812900244
|
5 years ago |
Megvii Engine Team
|
a1877ee0fa
|
refactor(dnn): refactor algo interface, use algoinfo instead of global algorithm
GitOrigin-RevId: 479718ac75
|
5 years ago |
Megvii Engine Team
|
6856ce9ce2
|
feat(dnn): support conv bias activation for nchw4 input tensor format and nchw output tensor format
GitOrigin-RevId: 29cd73f87b
|
5 years ago |
Megvii Engine Team
|
c03249c059
|
feat(dnn/opr): add megdnn fake quant opr
GitOrigin-RevId: 5a04b6da2f
|
5 years ago |
Megvii Engine Team
|
1217801133
|
perf(mge): add opdef for broadcast
GitOrigin-RevId: 92f0af29eb
|
5 years ago |
Megvii Engine Team
|
2a3f4d099a
|
refactor(dnn/arm): refactor CPU heuristic algo selection
GitOrigin-RevId: 60d2646bb3
|
5 years ago |
Megvii Engine Team
|
ba66e1d039
|
feat(dnn): add nchw_fp32 nchw44_qint8 cuda dct
GitOrigin-RevId: 581e31fc20
|
5 years ago |
Megvii Engine Team
|
215f88f373
|
fix(dnn/argmxx): fix argmxx on inf
GitOrigin-RevId: 740f67b73a
|
5 years ago |
Megvii Engine Team
|
edb32495c6
|
feat(dnn/opr): add megdnn adaptive pooling opr
GitOrigin-RevId: 563ce65479
|
5 years ago |
Megvii Engine Team
|
95eb6ae380
|
feat(mgb/opr): let more ops support empty IO
GitOrigin-RevId: 84dddb4b23
|
5 years ago |
Megvii Engine Team
|
a5fad7d07c
|
feat(dnn): add compile for riscv64
GitOrigin-RevId: fa0c163527
|
5 years ago |
Megvii Engine Team
|
3e11d89415
|
fix(dnn/dump): add more info for dump CD4
GitOrigin-RevId: 5840afaacd
|
5 years ago |
Megvii Engine Team
|
16324e3076
|
feat(dnn/cuda): add remap backward
GitOrigin-RevId: 1b1bcf5db3
|
5 years ago |
Megvii Engine Team
|
6e882c1a86
|
feat(whl/imperative): compat for build python whl imperative and legacy runtime
GitOrigin-RevId: 7f6629ae1f
|
5 years ago |
Megvii Engine Team
|
7f857bd471
|
feat(mgb/rocm): add cmake for rocm and fix compile errors and bn
GitOrigin-RevId: c73ed4adc3
|
5 years ago |
Megvii Engine Team
|
199eefbd4c
|
fix(dnn): generate mode files
GitOrigin-RevId: 9b1e840f00
|
5 years ago |
Megvii Engine Team
|
9510136223
|
fix(mgb/rocm): remove begin-internal of rocm
GitOrigin-RevId: 1523833fcb
|
5 years ago |
Megvii Engine Team
|
00ef677249
|
fix(mgb): remove internal for cambricon and atlas
GitOrigin-RevId: 861e349eb4
|
5 years ago |
Megvii Engine Team
|
a1e6720756
|
feat(dnn): enable bool comparison
GitOrigin-RevId: 735693b81e
|
5 years ago |
Megvii Engine Team
|
56381f808b
|
fix(dnn/arm): use vcvtq_f32_s32 for all arm code
GitOrigin-RevId: 27effe7d24
|
5 years ago |
Megvii Engine Team
|
1173205726
|
fix(gopt): nchw_nchwxx useable and opt pass use nchw_nchwxx_valid
GitOrigin-RevId: 60942aca5b
|
5 years ago |
Megvii Engine Team
|
2272abe18d
|
fix(mgb/fallback): disable nchw44 in conv1x1 and im2col in x86
GitOrigin-RevId: 603d2eb94a
|
5 years ago |
Megvii Engine Team
|
230ab45a1e
|
fix(mgb/naive): fix naive convolution no dispatch kernel in handle
GitOrigin-RevId: 4038fe23a4
|
5 years ago |
Megvii Engine Team
|
6e70fa7a11
|
feat(dnn/arm): add fp32 asm gemm for a53 a55 and i8i8i16 gemm for a72 a53
GitOrigin-RevId: a049c33f2b
|
5 years ago |
Megvii Engine Team
|
c7b6ef35c1
|
feat(dnn/cuda): add warp perspective backward mat idx
GitOrigin-RevId: b4b494bb69
|
5 years ago |
Megvii Engine Team
|
e258812f12
|
feat(dnn): add bool dtype
GitOrigin-RevId: 98c8a092b4
|
5 years ago |
Megvii Engine Team
|
6bcc6faec8
|
feat(mge/imperative/opr): modify batch_norm to support frozen BN
fix(mge/imperative): cmake uses MGE_BUILD_IMPERATIVE_RT flag
GitOrigin-RevId: 8ea21af9da
|
5 years ago |
Megvii Engine Team
|
f6018422fd
|
perf(dnn/arm_common): add nchw44 winograd f73
GitOrigin-RevId: 8ed98ab85b
|
5 years ago |
Megvii Engine Team
|
324af87807
|
feat(dnn/arm): add cpuinfo runtime check for x86 and arm
GitOrigin-RevId: c2020a344e
|
5 years ago |