Commits

Branch selector

 main

User selector

All users

Datepicker

All time

Commit History

Commits on Aug 18, 2022

• Qlinearsoftmax take FLOAT lookup-table (microsoft#12574)

  Show description for 81b128b
  wejoncy

  authored
  81b128b

Commits on Aug 17, 2022

• [oneDNN] Improve DequantizeLinear operator performance. (microsoft#12611)

  Show description for 82b724f
  eralmual

  authored
  82b724f

• Add BuildError for --gen_doc and --enable_training (microsoft#12630)

  

  Thiago Crepaldi

  authored
  d1ba801

• Replace to lock_guard as lighter class for locking (microsoft#12616)

  Show description for 9481893
  yuslepukhin

  authored
  9481893

• weight matching (microsoft#12607)

  Show description for f2db6bb
  chenfucn

  authored
  f2db6bb

• Fix a build error (microsoft#12600)

  Show description for 8a038b9
  hmchen-github

  authored
  8a038b9

• Improve LongformerAttention performance: AddBiasTranspose and New weight format (microsoft#12448)

  Show description for ce01ed0
  tianleiwu

  authored
  ce01ed0

• Refactor with std::variant (on device training) (microsoft#12383)

  Show description for 7df2e8c
  pengwa

  authored
  7df2e8c

Commits on Aug 16, 2022

• Replace references to onnxruntime 'master' with 'main' in Dockerfiles. (microsoft#12550)

  Show description for caabfcd
  edgchen1

  authored
  caabfcd

• Add build option to link TensorRT prebuilt parser (microsoft#12602)

  Show description for 9d10bad
  yf711

  authored
  9d10bad

• [Java] JNI refactor for OrtSession (microsoft#12496)

  Show description for 733db31
  Craigacp

  authored
  733db31

• QDQ debugger - activations compare (microsoft#12544)

  Show description for eb6aa86
  chenfucn

  authored
  eb6aa86

Commits on Aug 15, 2022

• release calibrator before deleting temporary files (microsoft#12601)

  yufenglee

  authored
  30ee5a4

• ppc64le: fix MlasQLinearMulKernel's VSX code to work with inputs of 32 bits (microsoft#12441)

  maxiwell

  authored
  19a9690

• ONNX Protobuf natvis with some google::protobuf (microsoft#12580)

  Show description for 6166771
  yuslepukhin

  authored
  6166771

• Add Learning Rate Scheduler C API (microsoft#11957)

  baijumeswani

  authored
  f5e3517

• Add TRT uint8 support (microsoft#12570)

  Show description for 73da3f3
  kevinch-nv

  authored
  73da3f3

Commits on Aug 12, 2022

• [js/web] update branch name for pull:wasm (microsoft#12548)

  Show description for 95f2a3e
  fs-eire

  authored
  95f2a3e

• Automate generation of javadocs and create PR with changes (microsoft#12515)

  natke

  authored
  cc9b3e1

• Support direct usage of ORT format model flatbuffer for initializers (microsoft#12465)

  Show description for 0b0c51e
  skottmckay

  authored
  0b0c51e

• Add FusedConv Op to ROCm (microsoft#11792)

  Show description for bc353c7
  xinyazhang

  authored
  bc353c7

• [ROCm] NGramRepeatBlock, LongformerAttention and DecoderAttention Ops (microsoft#11971)

  Show description for eb827bd
  xinyazhang

  authored
  eb827bd

Commits on Aug 11, 2022

• do not quantize Relu/Clip if their inputs are not quantized (microsoft#12565)

  yufenglee

  authored
  95df5da

• DirectML GEMM broken in opset 11 and 13 when optional tensor C not provided (microsoft#12568)

  Show description for 67f6b7c
  smk2007

  authored
  67f6b7c

• Adding w_zero_point to conv_integer_test.cc (microsoft#12423)

  Show description for 580f229
  jchen351

  authored
  580f229

• Updating binary ops in eager mode to support broadcasting. (microsoft#12560)

  Show description for 3d009cd
  WilBrady

  authored
  3d009cd

• Enable PythonOp for --enable_training_torch_interop build (microsoft#12539)

  Show description for 24eab92
  pengwa

  authored
  24eab92

• Add big endian support to murmurhash3 (microsoft#12549)

  skottmckay

  authored
  b59ccbc

• Fix Compile Warning (microsoft#12552)

  Show description for 018fba9
  centwang

  authored
  018fba9

• Remove CUDA 10.2 support (microsoft#12541)

  snnn

  authored
  ac7538b

• [xnnpack] basic QDQ operators support (microsoft#11912)

  Show description for 819c367
  

  wejoncy

  and

  Jicheng Wen

  authored
  819c367

Commits on Aug 10, 2022

• Add win-ci pipeline for on-device training (microsoft#12513)

  baijumeswani

  authored
  3e78f3c

• fix qdq relu removal bug (microsoft#12542)

  Show description for b2382dc
  chenfucn

  authored
  b2382dc

• Use alignas instead of naive padding to avoid false cache sharing (microsoft#12514)

  Show description for c10704a
  yuslepukhin

  authored
  c10704a

• Add default to TRT datatype switch statement (microsoft#12533)

  Show description for 25032f1
  kevinch-nv

  authored
  25032f1

Pagination

PreviousNext