summaryrefslogtreecommitdiff
path: root/caffe2
AgeCommit message (Expand)AuthorFilesLines
2019-04-23optimize BatchMatmulOp (#18612)Xiaomeng Yang6-406/+533
2019-04-23caffe2 | Windows compat fixesOleg Bogdanov2-40/+48
2019-04-23correct comments in group_norm_op (#19621)Huamin Li2-4/+4
2019-04-23Surface the Glow traces to C2 (#19087)Yinghai Lu2-5/+48
2019-04-23Allow extracting element-wise loss in softmax (#19579)Priya Goyal2-2/+4
2019-04-23Specify to use Float16UniformFill if necessary in sparse lookup layer (#18499)Jiyan Yang1-2/+7
2019-04-22Automatic update of fbcode/onnx to 0e8d2bc5e51455c70ef790b9f65aa632ed9bc8a7 (...Lu Fang1-0/+2
2019-04-22Add back option to not adjust output batch size (#19442)Yinghai Lu6-12/+38
2019-04-22Add debug logic to c2_ref_test and its helpers (#19359)Michael Antonov3-7/+27
2019-04-22fix variable shadowing issus (#19567)Dehua Cheng1-2/+2
2019-04-22Automatic update of fbcode/onnx to 83dd62659fc07d5b7fa93b5d1c1879f93509c7db (...Lu Fang1-0/+5
2019-04-21Add assertion to make sure init op is always fp16 compatible in fp16 trainingJiyan Yang1-9/+20
2019-04-20Improve optimizations for DNNLOWP support on MKL-DNN (#18843)Gu, Jinghui2-195/+740
2019-04-19Support compilation on gcc-7.4.0 (#19470)Sam Leeman-Munk1-0/+1
2019-04-19Improve embedding_bag add kernel (#19329)James Reed1-0/+3
2019-04-19Fix relu bug for empty tensor (#19451)Xiaomeng Yang2-6/+35
2019-04-19Fix out-of-topological-order issue in Nomnigraph (#19458)Yinghai Lu3-14/+32
2019-04-18Moving at::Tensor into caffe2::Tensor without bumping refcount (#19388)Sebastian Messmer1-2/+2
2019-04-18Use string based schema for exposing caffe2 ops (#19287)Sebastian Messmer8-115/+88
2019-04-17Add validator for optimizers when parameters are sharedJiyan Yang3-1/+153
2019-04-17Remove unused template parameter in OnnxifiOp (#19362)Yinghai Lu2-6/+6
2019-04-17Eliminate AdjustBatch ops (#19083)Yinghai Lu7-384/+202
2019-04-17Delete C10Tensor (#19328)Sebastian Messmer21-72/+49
2019-04-16Testing for folded conv_bn_relu (#19298)Jerry Zhang2-0/+47
2019-04-15Avoid undefined symbol error when building AdIndexer LTO (#19009)Mark Santaniello2-81/+81
2019-04-15Add NHWC order support in the cost inference function of 3d conv (#19170)Summer Deng1-9/+19
2019-04-15unit test with multiple op invocations (#19118)Jongsoo Park11-122/+117
2019-04-15Fix the return value of ParseFromString (#19262)Gemfield1-0/+1
2019-04-12Add more debugging helper to net transformer (#19176)Yinghai Lu3-19/+32
2019-04-12use C10_REGISTER for GELU opHuamin Li4-24/+48
2019-04-11Change is_variable() to check existence of AutogradMeta, and remove is_variab...Will Feng1-2/+1
2019-04-11Materialize a non-default device for C2 legacy storage. (#18605)Gregory Chanan1-2/+6
2019-04-11Allow empty net type (#19154)Yinghai Lu1-1/+1
2019-04-11try to enable uncertainty for lr loss (#17236)Xing Wang2-0/+56
2019-04-10Optimize SoftmaxOp on CPU (#18635)Xiaomeng Yang8-166/+138
2019-04-10Move ConcatBatchMatMulBatchGatherOp to OSSHao Lu2-0/+156
2019-04-10implement operators for DNNLOWP (#18656)Gu, Jinghui39-87/+1577
2019-04-10Clear input/ouput shape cache for each inference (#19085)Yinghai Lu1-0/+2
2019-04-09amend D14778810 (#18902)Summer Deng5-9/+36
2019-04-09Fix aten op output assignment (#18581)Wanchao Liang2-28/+41
2019-04-09Make BlackBoxPredictor handle networks throwing exceptions (#19080)Alexander Sidorov1-0/+9
2019-04-09remove interned_string.h dep (#19061)Lu Fang1-2/+1
2019-04-09add logging to make the saving action visible (#19042)Liang Xiong1-0/+2
2019-04-09Convert all tabs to spaces, add CI. (#18959)Edward Yang3-17/+17
2019-04-08Add gelu op (#18992)Xiaomeng Yang7-9/+392
2019-04-08Export C10 operator in PyTorch Model (#18210)Lu Fang2-1/+6
2019-04-08caffe2 - Expose tensor filler util to Python (#18886)Duc Ngo9-1/+95
2019-04-08Fix a dev mode bug in activation distribution observer (#19004)Summer Deng2-3/+9
2019-04-07Remove tensorFromBlob() from Type (#18779)Roy Li1-6/+10
2019-04-06fix bug when falling back to acc32 when weight is prepacked (#18974)Summer Deng4-122/+213