合并请求 · PaddlePaddle / Paddle

Enable MKL-DNN INT8 Concat Kernel.

!16156 · opened 3月 12, 2019 by saxon_zh Intel int8
- 已合并
- 24
更新于3月 26, 2019
Add cpu_quantize_squash_pass for C-API quantization

!16128 · opened 3月 08, 2019 by saxon_zh Intel int8
- 已合并
- 2
更新于3月 14, 2019
Add cpu_quantize_pass for C-API quantization

!16127 · opened 3月 08, 2019 by saxon_zh Intel int8
- 已合并
- 14
更新于3月 18, 2019
resolve #15618

!16114 · opened 3月 08, 2019 by saxon_zh Intel
- 已合并
- 4
更新于3月 13, 2019
[Testing PR] FC with disabled components (GPU,Anakin,NGRAPHS etc.) . (DO NOT MERGE and DO NOT REVIEW)

!16108 · opened 3月 07, 2019 by saxon_zh Intel
- 已关闭
- 0
更新于3月 12, 2019
[Testing PR]No fcmkldnn (no review & no merge)

!16091 · opened 3月 06, 2019 by saxon_zh Intel
- 已关闭
- 0
更新于3月 28, 2019
Add PaddleDType INT32 support

!16090 · opened 3月 06, 2019 by saxon_zh Intel
- 已合并
- 4
更新于3月 08, 2019
Fix the node's order issue when the content of graph is changed

!16088 · opened 3月 06, 2019 by saxon_zh Intel
- 已合并
- 5
更新于3月 08, 2019
[JUST TESTING CI] Testing PR (no merge & no review)

!16064 · opened 3月 05, 2019 by saxon_zh Intel
- 已关闭
- 1
更新于3月 28, 2019
fix typo on tanh_grad op

!16049 · opened 3月 05, 2019 by saxon_zh Intel NGraph
- 已合并
- 0
更新于3月 05, 2019
MKLDNN: Add UT for conv_transpose_mkldnn op.

!16030 · opened 3月 03, 2019 by saxon_zh Intel
- 已合并
- 0
更新于3月 05, 2019
fix cpplint in ngraph file

!16028 · opened 3月 03, 2019 by saxon_zh Intel NGraph
- 已合并
- 0
更新于3月 04, 2019
MKLDNN: Add UT for conv2d_mkldnn_op with fuse_bias and fuse_residual

!16016 · opened 3月 01, 2019 by saxon_zh Intel
- 已合并
- 0
更新于3月 04, 2019
MKL-DNN: Add a unit test for pooling when ceil mode is enabled

!16013 · opened 3月 01, 2019 by saxon_zh Intel
- 已合并
- 0
更新于3月 02, 2019
C-API quantization core

!15987 · opened 3月 01, 2019 by saxon_zh Intel int8
- 已关闭
- 71
更新于3月 29, 2019
added concat op for ngraph engine

!15946 · opened 2月 27, 2019 by saxon_zh Intel NGraph
- 已合并
- 0
更新于2月 27, 2019
MKL-DNN: Add placement pass tester

!15943 · opened 2月 26, 2019 by saxon_zh Intel
- 已合并
- 0
更新于2月 27, 2019
The flag of mkldnn engine is enabled iff it is necessary

!15941 · opened 2月 26, 2019 by saxon_zh Intel
- 已合并
- 0
更新于2月 28, 2019
MKLDNN: Add UT for conv2d_mkldnn_op with fuse_bias and fuse_residual

!15936 · opened 2月 26, 2019 by saxon_zh Intel
- 已关闭
- 5
更新于3月 01, 2019
Optimize gelu operation with mkl erf

!15931 · opened 2月 26, 2019 by saxon_zh Intel
- 已合并
- 4
更新于2月 28, 2019

PaddlePaddle / Paddle 大约 1 年 前同步成功

PaddlePaddle / Paddle
大约 1 年前同步成功