summaryrefslogtreecommitdiff
path: root/caffe2
AgeCommit message (Expand)AuthorFilesLines
2019-02-28add dropout during eval (#17549)Huan Gui2-16/+30
2019-02-28Allow exposing caffe2 operators with variable number of input tensors to c10 ...Sebastian Messmer2-11/+30
2019-02-28Revert D14078519: [codemod][caffe2] [clangr] refactor caffe2 operator constru...Sebastian Messmer25-142/+106
2019-02-28refactor caffe2 operator constructors - 8/9 (#17089)Sebastian Messmer25-93/+137
2019-02-28refactor caffe2 operator constructors - 6/9 (#17087)Sebastian Messmer25-68/+86
2019-02-28refactor caffe2 operator constructors - 2/9 (#17083)Sebastian Messmer25-142/+192
2019-02-28refactor caffe2 operator constructors - 7/9 (#17088)Sebastian Messmer25-111/+136
2019-02-28refactor caffe2 operator constructors - 4/9 (#17085)Sebastian Messmer25-102/+142
2019-02-28refactor caffe2 operator constructors - 3/9 (#17084)Sebastian Messmer25-127/+153
2019-02-28refactor caffe2 operator constructors - 5/9 (#17086)Sebastian Messmer25-106/+142
2019-02-28ONNXIFI extension & e2e tests. (#17478)Rui Zhu2-34/+85
2019-02-28refactor caffe2 operator constructors - 9/9 (#17090)Sebastian Messmer7-33/+46
2019-02-27Remove useless OpenCV referenceDmytro Dzhulgakov3-12/+0
2019-02-27Make C10_MOBILE consistent with how feature macros are usually used (#17481)Sebastian Messmer13-30/+18
2019-02-27Disable c10 dispatcher on mobile (#17078)Sebastian Messmer2-0/+17
2019-02-27Fix math::Set for large tensor (#17539)Xiaomeng Yang4-19/+21
2019-02-27optimize elementwise sum (#17456)Jongsoo Park2-61/+136
2019-02-27Enable boolean_mask, adadelta, adagrad fp16 on ROCm (#17235)rohithkrn4-17/+17
2019-02-26Fix operator initialization order (#15445)Ilia Cherniavskii1-2/+4
2019-02-25Back out part of "Fix NERPredictor for zero initialization"Dmytro Dzhulgakov1-5/+5
2019-02-25Automatic update of fbcode/onnx to e18bb41d255a23daf368ffd62a2645db55db4c72 (...Lu Fang1-0/+1
2019-02-25Apply modernize-use-override (4)Michael Liu1-2/+2
2019-02-25Export ElementwiseLinear to ONNX (Mul + Add). (#17411)Tongliang Liao4-5/+112
2019-02-25Add foxi submodule (ONNXIFI facebook extension)Lu Fang2-2/+2
2019-02-22remove redundant inference functions for FC (#17407)Jongsoo Park3-72/+21
2019-02-22optimize max pool 2d (#17418)Jongsoo Park4-21/+142
2019-02-22Better handling of net errors in prof_dag counters (#17384)Ilia Cherniavskii3-26/+102
2019-02-22Fix install libcaffe2_protos.a issue mentioned in #14317 (#17393)Gemfield1-0/+3
2019-02-22Improve onnxifi backend init time (#17375)Yinghai Lu4-64/+54
2019-02-22Revert D14181620: [caffe2/int8] optimize max pool 2dLu Fang3-141/+21
2019-02-22fallback operators to CPU for onnx support (#15270)Gu, Jinghui2-8/+71
2019-02-22optimize max pool 2d (#17391)Jongsoo Park3-21/+141
2019-02-21Fix concat dimension check bug (#17343)Chandler Zuo1-1/+4
2019-02-20Skip convnets benchmark in rocm CI (#17331)Junjie Bai1-1/+2
2019-02-20Implementation convolutionTranspose operator for mkl-dnn (#12866)Cheng,Penghui4-9/+587
2019-02-20Support multi-device configuration for MKL-DNN (#12856)Cheng,Penghui1-3/+16
2019-02-20fix missing std (#17263)Ailing Zhang1-1/+1
2019-02-20Rethrow exceptions from RunAsync (#15034)Ilia Cherniavskii1-1/+5
2019-02-20Reinforce scheduling invariants (#17132)Ilia Cherniavskii4-124/+147
2019-02-20Modify TileOp GPU implementation to expose more concurrency and better utiliz...Lukasz Wesolowski1-15/+17
2019-02-20Separate gpu reduce functions (#17146)Xiaomeng Yang6-788/+698
2019-02-20Removed obsolete argument correct_transform_coords in bbox_transform op. (#16...Peizhao Zhang8-63/+20
2019-02-20make the threshold for acurracy more precise (#17194)Hector Yuen1-13/+23
2019-02-20Add rule based filtering for ONNXIFI transformation (#17198)Yinghai Lu6-212/+331
2019-02-20caffe2 | added missing operator source file (#17272)Oleg Bogdanov1-0/+1
2019-02-19Fix dll loading issue for Caffe2 and Windowspeter1-5/+23
2019-02-17Logging stuffs (#17177)Ying Zhang1-0/+8
2019-02-16Skip onnx logsoftmax tests in rocm (#17170)Junjie Bai1-0/+1
2019-02-15change the epsilon for fp32/fp16 to uint8 to be the same (#17062)Hector Yuen2-29/+75
2019-02-15Fix batch insert (#17158)Yinghai Lu4-30/+43
generated by cgit v1.2.3 (git 2.25.1) at 2024-11-27 04:11:01 +0000