summaryrefslogtreecommitdiff
path: root/caffe2/operators/quantized
AgeCommit message (Expand)AuthorFilesLines
2019-02-06caffe2 | MSVS compatibility fixes (#16765)Oleg Bogdanov6-1/+27
2019-01-30Back out "[pt1][tensor] Change ConvPoolOp<Context>::SetOutputSize to ConvPool...Jerry Zhang3-6/+3
2019-01-28Change SetOutputSize in ConvTransposeUnpoolBaseOp (#16179)Jerry Zhang1-1/+2
2019-01-28Change ConvPoolOp<Context>::SetOutputSize to ConvPoolOp<Context>::GetOutputSi...Jerry Zhang3-3/+6
2019-01-16Update int8_simd.h (#13859)JerryShih1-72/+1
2019-01-16Tensor reinitialization codemod - 3/5 (#15912)Jerry Zhang2-2/+2
2019-01-11Tensor reinitialization codemod - 4/5 (#15967)Jerry Zhang5-12/+19
2018-12-27Update QNNPACK (#15561)Marat Dukhan11-2/+13
2018-12-13Tensor construction codemod(ResizeLike) - 6/7 (#15137)Jerry Zhang1-2/+2
2018-12-11Switch Int8Softmax, Int8Relu, and Int8LeakyRelu to QNNPACK (#14933)Marat Dukhan3-192/+200
2018-12-08Switch Int8Sigmoid to QNNPACK (#14883)Marat Dukhan1-193/+72
2018-12-06Switch Int8MaxPool operator to QNNPACK (#14832)Marat Dukhan1-140/+55
2018-12-05Validate matching input shapes in Int8Add operator (#14520)Marat Dukhan1-0/+5
2018-12-05Switch Int8AveragePool operator to QNNPACK (#14783)Marat Dukhan1-159/+110
2018-11-26Switch Int8ChannelShuffle operator to QNNPACK (#14362)Marat Dukhan1-117/+53
2018-11-18Switch Int8Add operator to QNNPACK (#14089)Marat Dukhan1-186/+73
2018-11-16Rename neon2sse.h to NEON_2_SSE.h to match upstream repoMarat Dukhan1-1/+1
2018-11-07operators/quantized/server -> quantization/server (#13660)Jongsoo Park116-21018/+0
2018-11-07caching packed matrix (#13626)Jongsoo Park5-12/+120
2018-11-07new QNNPACK dwconv support and tests (#13652)Yiming Wu1-0/+122
2018-11-07swap with empty vector to force deallocation (#13625)Jongsoo Park5-5/+5
2018-11-07Rename ndim() -> dim() - 4/6Jerry Zhang10-23/+22
2018-11-06Remove C2GEMMContext (#13443)Marat Dukhan6-79/+16
2018-11-06Rename ndim() -> dim() - 5/6Jerry Zhang1-2/+2
2018-11-06CMake integration for int8 server operatorsDaya S Khudia35-117/+423
2018-11-05Rename dim(i) -> size(i) - 1/2Jerry Zhang1-2/+2
2018-11-05Add Int8BatchPermutation op in DNNLOWP (#13539)Viswanath Sivakumar3-0/+140
2018-11-04Duplicate bias blobs shared by different conv ops to handle scale correctly (...Viswanath Sivakumar1-0/+40
2018-11-04remove unnecessary files (#13537)Jongsoo Park2-251/+0
2018-11-02Reslove lint and infer warning (#13520)Haixin Liu6-9/+8
2018-11-02Rename dim(i) -> size(i) - 2/2Jerry Zhang6-43/+43
2018-11-02Migrate dnnlowp code to open source directory (#13500)Haixin Liu112-0/+20677
2018-10-31int8 sigmoid op (#13298)Yiming Wu7-35/+346
2018-10-30Renaming meta() to dtype() - 2/2 (#13334)Jerry Zhang2-3/+3
2018-10-29Renaming size() to numel() - 1/2Jerry Zhang14-45/+45
2018-10-27Fix performance regresion introduced in D10524381 (#13199)Marat Dukhan6-1/+5
2018-10-25CMake integration for Int8 opsMarat Dukhan1-0/+31
2018-10-25Open-source Caffe2 Int8 ops (#13065)Marat Dukhan44-0/+4540
generated by cgit v1.2.3 (git 2.25.1) at 2024-11-23 04:59:48 +0000