root
/
MNN
mirror of https://github.com/alibaba/MNN.git
1
0
Fork 0
Code Issues Actions 1 Packages Projects Releases Wiki Activity
1,034 Commits 20 Branches 109 Tags 396 MiB
Commit Graph

214 Commits

Author SHA1 Message Date
tianhang.yth 4eb1096b9c fix alpha div by zero bug and arm server compile bug 2021-06-24 10:38:55 +08:00
Joker 4184860ae4 feat(arm82): add GridSample op support in arm82 backend 2021-06-23 14:10:31 +08:00
xiaying f6422c315c [MNN:Bugfix] Fix bug for ConvInt8TiledExecutor onClone 2021-06-16 16:20:42 +08:00
xiaying 8d9f86bc4a fix compile bug for c++ < 14 2021-06-16 15:24:46 +08:00
hush-alibaba 58545d6ca1
Synchronize internal github for version 1.2.0 (#1518) 2021-06-11 17:17:13 +08:00
tianhang.yth d85952d826 sync from internal repo 2021-04-28 18:02:10 +08:00
xiaying b62c2eb687 [BF16:Bugfix] Fix compile bug for BF16 in NO SSE and NO NEON 2021-04-21 15:54:01 +08:00
xiaying c2a2a24e8e [IOS:Bugfix] Fix compile bug for IOS Demo 2021-04-21 15:01:56 +08:00
xiaying 3c4ba7c595 [MNN:Sync] Sync internal gitlab 2021-04-16 14:50:43 +08:00
xiaying 3e06cabf38 [PATCH 22/36] Fix bug for CPUScatterNd crash for invalid input 2021-04-16 14:29:37 +08:00
弗人 95bcb842a0 [PATCH 08/36] [Train:Feature:Bugfix] train quant support full quant 2021-04-16 14:29:36 +08:00
jason_w 29a7128bdb
Remove unnecessary '+' 
Remove unnecessary '+'
 2021-04-13 08:12:38 +08:00
xiaying d91fc63976 [MNN:Sync] Sync internal Gitlab 2021-04-08 15:34:23 +08:00
xiaying 7af23d29f4 [MNN:Bugfix] Fix compile bug 2021-04-01 13:57:02 +08:00
xiaying 1181033ca4 Fix onnx scatternd convert bug 2021-03-31 11:53:41 +08:00
xiaying 1c4eded014 [MNN:Bugfix] Fix bug for CPUPool for w, h < kernelSize and pad 2021-03-04 22:46:06 +08:00
xiaying 5e127496fc Sync Internal Github 2021-02-07 10:47:03 +08:00
jokerz0624 bfc91d2b87
fix(SSE): fix one bug of PRelu in function _SSE_MNNReluWithSlopeChannel (#1371) 2021-02-07 10:35:47 +08:00
Joker 3a121b71ff improvement(arm82): accelerate data format converting between NC8HW8 and NCHW in armv82 2021-01-30 16:06:59 +08:00
xiaying d766c6d7c0 Fix bug for compute MatMul E=1 for multi-thread 2021-01-11 14:04:25 +08:00
xiaying 4b1a11d9dc [PATCH 14/19] [Arm82:Feature] Support All Binary of float for arm82 2021-01-08 14:16:25 +08:00
xiaying 7f94e02410 [PATCH 04/19] [Converter:Bugfix] Support group convolution for PB 2021-01-08 14:16:24 +08:00
xiaying 2d1b129121 [MNN:Sync] Sync internal git 2021-01-06 16:29:37 +08:00
xiaying aedc8f6a68 [PATCH 341/350] Add avx512 patch 2021-01-06 15:57:22 +08:00
xiaying 83f577fbfb [PATCH 335/350] [MNN:Bugfix] Fix bug for int8 sse not align 2021-01-06 15:57:22 +08:00
xiaying 49b0544719 [PATCH 314/350] [MNN:Bugfix] Fix memory usage bug for OneDNNConvolution 2021-01-06 15:57:20 +08:00
xiaying f258d6bd92 [PATCH 279/350] [MNN:Bugfix] Fix comiple bug for cross-compile of ARM 2021-01-06 15:57:17 +08:00
xiaying 0fe2b0dfee [PATCH 278/350] [MNN:Speed] Support OneDNN for MNN Convolution 2021-01-06 15:57:17 +08:00
弗人 e6697960f0 [PATCH 276/350] bugfix for topk neon compile 2021-01-06 15:57:17 +08:00
弗人 d3d34f16f1 [PATCH 275/350] [MNN:Bugfix] bugfix for topk neon compile 2021-01-06 15:57:16 +08:00
和彬 c328e8ff15 [PATCH 270/350] fix conv depthwise overflow bug 2021-01-06 15:57:16 +08:00
雁行 63acf0e8a9 [PATCH 269/350] [MNN:Bugfix] fix bug of Pool pads 2021-01-06 15:57:16 +08:00
houjiang ef942629ec [PATCH 257/350] [MNN::Bugfix] Fix avx load statments. 2021-01-06 15:57:15 +08:00
houjiang e3262eac4c [PATCH 243/350] [MNN::Refine] Refine code style. 2021-01-06 15:57:14 +08:00
houjiang 6b76f54f94 [PATCH 241/350] [MNN::Bugfix] Explicitly specify align boundary. 2021-01-06 15:57:14 +08:00
xiaying 2d96127417 [PATCH 225/350] [MNN:Bugfix] Set GatherV2 as the same with onnx runtime 2021-01-06 15:57:13 +08:00
xiaying 3d51e1454d [PATCH 218/350] Fix bug for gcc compute log slow 2021-01-06 15:57:12 +08:00
xiaying 9a51854399 [PATCH 217/350] Optimize convInt8Remain for e = 1 2021-01-06 15:57:12 +08:00
houjiang 9c7bbc4fe9 [PATCH 216/350] [MNN::Refine] Make the weight align boundary-independent by redefining x86 instructions. 2021-01-06 15:57:12 +08:00
houjiang d7d7ece8f4 [PATCH 214/350] [MNN::Refine] Rearrange weights for 1x1 and generic convolution. 2021-01-06 15:57:12 +08:00
houjiang 5cb6071469 [PATCH 203/350] [MNN::Feature] Rearrange weights. 2021-01-06 15:57:11 +08:00
弗人 74fd76e72c [PATCH 201/350] [MNN:Feature] add support for sparse+convint8 2021-01-06 15:57:11 +08:00
xiaying b18216a614 [PATCH 198/350] Support one-scale float2int8/int8tofloat 2021-01-06 15:57:10 +08:00
DaydreamCoding 5a9d779237 [PATCH 196/350] fix threadpool may not destroyed 2021-01-06 15:57:10 +08:00
xiaying 57d67dde2d [PATCH 191/350] Speed up Int8Tofloat for x86-sse 2021-01-06 15:57:10 +08:00
xiaying a131c8cd26 [PATCH 188/350] [MNN:Speed] Optimize ConvInt8 for remain case 2021-01-06 15:57:09 +08:00
xiaying a943d45796 [PATCH 187/350] Float2Int8 opt for x86-sse 2021-01-06 15:57:09 +08:00
Hui Shu ab711d484c Synchronize internal master to Github 2020-12-15 14:12:35 +08:00
DaydreamCoding d4316ada8f
fix threadpool may not destroyed 2020-12-03 15:25:45 +08:00
雁行 08d6ae5ee7 [PATCH 70/78] [MNN:Bugfix] fix bug of ArgMax whith topk and value 2020-11-25 18:57:56 +08:00