Welcome to mirror list, hosted at ThFree Co, Russian Federation.

github.com/marian-nmt/FBGEMM.git - Unnamed repository; edit this file 'description' to name the repository.
summaryrefslogtreecommitdiff
AgeCommit message (Expand)Author
2020-05-10Add prepacked B matrixyouki/upstream0509Young Jin Kim
2020-02-07clean up temporary code for isa-dependent serialized packed weight (#290)Jongsoo Park
2020-02-07row-wise sparse fp32/fp16/int8 EmbeddingSpMDM (#288)Jongsoo Park
2020-02-05enable github action build/test on windows (#285)Hongzhang Shan
2020-02-05changes for shared release on windows (#284)Hongzhang Shan
2020-02-04enable positional weight for EmbeddingSpMDMNBitRowWiseSparse (#281)Jongsoo Park
2020-02-04enable EmbeddingSpMDMNBitRowWiseSparse avx2 (#280)Jongsoo Park
2020-02-02Fix vector size setting and min max set for randFill (#283)Hongzhang Shan
2020-02-02fix "invalid min and max arguments for uniform_int" error (#282)Hongzhang Shan
2020-02-01fix overflow in cache flush (#260)Jongsoo Park
2020-02-01add 2-stage SparseAdagrad interface (#276)Jiecao Yu
2020-02-01clang-format the code (#278)Jianyu Huang
2020-02-01Fix the root cause for _aligned_free issues on row_offset_ allocation (#277)Jianyu Huang
2020-02-01put cmake cpuinfo vars into cmake cacheHongzhang Shan
2020-01-31Fix the _aligned_free() issue on Windows (#275)Jianyu Huang
2020-01-31add more instantiation for EmbeddingSpMDMAvx2 (#274)Hongzhang Shan
2020-01-312/4-bit EmbeddingSPMDM with rowwise sparsity (#263)Jongsoo Park
2020-01-31Enable FMA operation for windows (#273)Hongzhang Shan
2020-01-312bit JIT'ed SLS kernel (#234)Jongsoo Park
2020-01-31cmake changes for windows compilingHongzhang Shan
2020-01-29Add instantiation for template parametersHongzhang Shan
2020-01-28Remove extra paranthesis (#270)Jianyu Huang
2020-01-28make gconv work for any W and H (#269)Jongsoo Park
2020-01-24use 2-stage EmbeddingSpMDM interface from benchmark and test (#256)Jongsoo Park
2020-01-23add missing copyright header (#267)Jongsoo Park
2020-01-22EmbeddingSpMDM with float16 input (#265)Jongsoo Park
2020-01-22modified instrinsic fp16 kernel for windows build (#259)Jongsoo Park
2020-01-22Intrinsic implementation of FP16 kernels for windows build (#254)Young Jin Kim
2020-01-22don't need extra register for positional weights (#261)Jongsoo Park
2020-01-22specialize fp32 SLS for emb dim 1 (#264)Jongsoo Park
2020-01-21avx2 4-bit SLS (#262)Jongsoo Park
2020-01-21remove EmbeddingSpMDM4BitKernelSignature (#257)Jongsoo Park
2020-01-19Add github action for windowsHongzhang Shan
2020-01-18enable positional weight test (#252)Jongsoo Park
2020-01-18Update readme with link to wiki pageDaya Khudia
2020-01-17Bug fixes in default path for fp16 (#251)Daya Khudia
2020-01-17add 2-stage EmbeddingSpMDM interface and change CodeCache to use read-write l...Jongsoo Park
2020-01-16pass the number of rows to EmbeddingSpMDM4Bit (#245)Jongsoo Park
2020-01-15Add FBGEMM_ENUM_CLASS_API (#249)Hongzhang Shan
2020-01-14Fix undefined reference error on macos (#247)Hongzhang Shan
2020-01-14Cmake changes for MSVC build (#242)Daya Khudia
2020-01-14Allocate memory instead of using variable length arrays (#241)Daya Khudia
2020-01-14More fixes related to MSVC build (#240)Daya Khudia
2020-01-14fix perf regression in resnet50 and resnet101 (#239)Daya Khudia
2020-01-13Fix multi-instance benchmarking for fp16 (#246)Daya Khudia
2020-01-10attributes not at the right placeDaya Khudia
2020-01-104-bit SLS with emb dim not a multiple of 2 (#243)Jongsoo Park
2020-01-09Githubaction (#237)Hongzhang Shan
2020-01-074bit JIT'ed SLS kernel (#233)Jongsoo Park
2020-01-05start to add github action fileHongzhang Shan