Commits

Branch selector

 master

User selector

ciyongch

Datepicker

All time

Commit History

Commits on Jan 9, 2020

• fix latency calculation and print issue (apache#17217)

  Show description for 4ff1c67
  

  ciyongch

  authored and

  pengzhao-intel

  committed
  4ff1c67

  Copy full SHA for 4ff1c67

Commits on Jan 3, 2020

• [MKLDNN] Support channel wise quantization for FullyConnected (apache#17187)

  Show description for 89fe1f6
  

  ciyongch

  authored and

  TaoLv

  committed
  89fe1f6

  Copy full SHA for 89fe1f6

Commits on Aug 26, 2019

• [MKLDNN] Fix out of bound access of req vector (apache#16000)

  Show description for 5042d2a
  

  ciyongch

  authored and

  TaoLv

  committed
  5042d2a

  Copy full SHA for 5042d2a

Commits on Aug 22, 2019

• [MKLDNN]Support fullyconnected and element-wise ops fusion (apache#15950)

  Show description for 434f185
  

  ciyongch

  authored and

  pengzhao-intel

  committed
  434f185

  Copy full SHA for 434f185

Commits on May 23, 2019

• [Quantization] Support zero-size tensor input for quantization flow (apache#15031)

  Show description for d4e458e
  

  ciyongch

  authored and

  pengzhao-intel

  committed
  d4e458e

  Copy full SHA for d4e458e

Commits on May 15, 2019

• Add primitive cache for MKL-DNN sum(elemwise_add operator (apache#14914)

  Show description for c5265fb
  

  ciyongch

  authored and

  pengzhao-intel

  committed
  c5265fb

  Copy full SHA for c5265fb

Commits on Apr 21, 2019

• Change size_t to int within for loop to fix windows build error (apache#14740)

  Show description for 100586a
  

  ciyongch

  authored and

  TaoLv

  committed
  100586a

  Copy full SHA for 100586a

Commits on Mar 27, 2019

• Performance improving for MKL-DNN Quantized FullyConnected (apache#14528)

  Show description for 5d2a451
  

  ciyongch

  authored and

  pengzhao-intel

  committed
  5d2a451

  Copy full SHA for 5d2a451

Commits on Mar 8, 2019

• MKLDNN based Quantized FullyConnected Operator and its fusion (apache#14128)

  Show description for 8668db7
  

  ciyongch

  authored and

  TaoLv

  committed
  8668db7

  Copy full SHA for 8668db7

Commits on Jan 15, 2019

• fix ssd quantization script error (apache#13843)

  Show description for 4fe5461
  

  ciyongch

  authored and

  TaoLv

  committed
  4fe5461

  Copy full SHA for 4fe5461

Commits on Dec 12, 2018

• fix quantize pass error when the quantization supported Op are excluded in the model (apache#13596)

  

  ciyongch

  authored and

  reminisce

  committed
  e36f888

  Copy full SHA for e36f888

Commits on Nov 30, 2018

• fix quantize_graph pass error when there're multiple outputs from a single node (apache#13000)

  Show description for 07a4319
  

  ciyongch

  authored and

  TaoLv

  committed
  07a4319

  Copy full SHA for 07a4319

Commits on Aug 18, 2018

• Add new test case for topk to cover large size of matrix as input (apache#12217)

  

  ciyongch

  authored and

  szha

  committed
  ca92e3d

  Copy full SHA for ca92e3d

Commits on Aug 16, 2018

• Fix the topk regression issue (apache#12197) (apache#12202)

  Show description for 1d31776
  

  ciyongch

  authored and

  szha

  committed
  1d31776

  Copy full SHA for 1d31776

Commits on Aug 13, 2018

• Accelerate the performance of topk for CPU side (apache#12085)

  Show description for 95dd95c
  

  ciyongch

  authored and

  szha

  committed
  95dd95c

  Copy full SHA for 95dd95c