summaryrefslogtreecommitdiff
path: root/caffe2/operators
AgeCommit message (Expand)AuthorFilesLines
2018-04-22Fix caffe2 eigen + cuda9 windows build (#6746)bddppq4-59/+59
2018-04-19[caffe2] Add ReduceMinOp and ReduceMaxOp (#6744)Xiaomeng Yang4-37/+443
2018-04-18ChannelShuffle with NHWC layout (#6667)Jongsoo Park3-28/+120
2018-04-18[caffe2] Update EigenTensorMap to use ColMajor (#6735)Xiaomeng Yang3-6/+0
2018-04-18[caffe2] Use both __ARM_NEON__ and __ARM_NEON macros (#6697)Marat Dukhan7-43/+43
2018-04-18[Caffe2] Enhance test for CollectAndDistributeOp (#6693)Pooya Davoodi1-1/+4
2018-04-18Add reduce min and reduce max (#6685)Xiaomeng Yang1-1/+1
2018-04-18[Caffe2] Detectron fpn support (#6645)daquexian1-36/+30
2018-04-17Update from Facebook (#6692)Orion Reblitz-Richardson6-12/+117
2018-04-17Add broadcast and reduce gradient (#6668)Xiaomeng Yang7-70/+150
2018-04-17Explicitly define all caffe2 reducer ops by name (#6513)costin-eseanu1-111/+215
2018-04-16Follow the change of ONNX Cast operator "to" attribute (#6574)bddppq1-9/+10
2018-04-14[caffe2] Update math functions with params on host. (#6602)Xiaomeng Yang11-351/+193
2018-04-11[caffe2] Update ReduceOps (#6497)Xiaomeng Yang9-349/+129
2018-04-11[Caffe2] Add support to TensorRT (#6150)Yinghai Lu1-0/+5
2018-04-10[Caffe2] Changes done inside Facebook (#6378)Yinghai Lu16-242/+1164
2018-04-09Guard couple shape inference functions for unkown input shapes (#6379)bddppq4-6/+21
2018-04-06Add missing attributes to the schema GivenTensorFill operators (#6330)bddppq1-0/+105
2018-04-04Remove eigen impl for arg_max and arg_min (#6293)Xiaomeng Yang2-52/+0
2018-04-02[caffe2] Fbcode to GitHub sync (#6208)Paul Jesse Hellemn1-10/+17
2018-03-30Revert D7394363: [GanH]: Log D Trick for Cross Entropy with SigmoidAndrey Malevich2-48/+5
2018-03-30extend bucketize op to support duplicated boundriesEllie Wen1-7/+17
2018-03-30CUDA version of LengthsMax operatorJason Gauci3-2/+389
2018-03-30[GanH]: Log D Trick for Cross Entropy with SigmoidXiaolong Wang2-5/+48
2018-03-30Fix schema check for arg_opsXiaomeng Yang1-3/+3
2018-03-30bugfix im2col opJay Mahadeokar1-2/+2
2018-03-30[easy] Improving error messageKittipat Virochsiri1-4/+5
2018-03-30[PerfModel] Added analytical counters for FCTransposed, BatchMatMul, BatchOneHotMartin Schatz4-94/+208
2018-03-30Back out "[PerfModel] Added analytical counters for FCTransposed, BatchMatMul...Martin Schatz4-208/+94
2018-03-27Move the stump op to ossLu Fang3-0/+156
2018-03-27Fix the CMake issues caused by internal changesLu Fang2-1/+40
2018-03-27Add SparseLengthsPositionalWeightedSum operator that fuses SparseLengthsWeigh...Jongsoo Park2-7/+47
2018-03-27fix lengths sum cuda op for empty batchXianjie Chen1-30/+57
2018-03-27fix cuda elementwise ops for empty batchXianjie Chen1-2/+1
2018-03-27Integrate fbgemm fp16 with Caffe2Summer Deng2-0/+63
2018-03-27[PerfModel] Added analytical counters for FCTransposed, BatchMatMul, BatchOneHotMartin Schatz5-95/+241
2018-03-27Add SumElementsInt operatorRoxie He3-0/+34
2018-03-27Faster positive modulo in IndexHashOpJanusz Kudelka1-2/+4
2018-03-27Back out "[PerfModel] Added analytical counters for FCTransposed, BatchMatMul...Martin Schatz5-249/+95
2018-03-27[PerfModel] Added analytical counters for FCTransposed, BatchMatMul, BatchOneHotMartin Schatz5-96/+249
2018-03-27Change observer copy() method to take id parameterMartin Schatz14-24/+79
2018-03-27Remove Apache headers from source.Orion Reblitz-Richardson490-7834/+0
2018-03-27Revert "[C2][GPU]LengthsMax CUDA version (#2209)" (#2444)Jason Gauci2-374/+1
2018-03-27[C2][GPU]LengthsMax CUDA version (#2209)harouwu2-1/+374
2018-03-26Caffe2-onnx exporter (#2248)Yinghai Lu5-5/+9
2018-03-23Add axis to top_k_op. (#2416)Xiaomeng Yang3-280/+506
2018-03-22Add InheritOnnxSchema property to c2 op schema (#2366)Yinghai Lu30-50/+96
2018-03-22Fix compiling issue with CAFFE2_NO_SANITIZE (#2386)Yinghai Lu1-1/+1
2018-03-22Add argmax and argmin ops (#2371)Xiaomeng Yang4-0/+565
2018-03-21Fix ReduceMean performance by specializing Eigen implementation for common sh...James Reed2-6/+160