PaddlePaddle / Paddle
大约 1 年前同步成功

代码
- 文件
- 提交
- 分支
- Tags
- 贡献者
- 分支图
- Diff
Issue 1423
- 列表
- 看板
- 标记
- 里程碑
合并请求 543
Wiki 0
- Wiki
分析
- 仓库
- DevOps
项目成员
Pages

elementwise_mul_grad 1.5比1.4慢很多

#19275 · opened 8月 19, 2019 by saxon_zh performance tuning
- 已关闭
- 1
更新于12月 30, 2019
模型结构里面出现“常数乘以op”的情况导致训练速度变慢

#19023 · opened 8月 06, 2019 by saxon_zh performance tuning
- 已关闭
- 4
更新于8月 10, 2020
[MKL-DNN] Improve performance of hashing Key genration using TID

#18902 · opened 7月 30, 2019 by saxon_zh Intel performance tuning
- 已关闭
- 2
更新于11月 12, 2019
[MKL-DNN] Extend reusing of primitives for Elementwise_add_mkldnn op

#18047 · opened 6月 12, 2019 by saxon_zh Intel performance tuning
- 已关闭
- 2
更新于7月 03, 2019
mpi训练速度还是很慢，已经使用py_reader读取数据

#15553 · opened 1月 28, 2019 by saxon_zh User performance tuning
- 已关闭
- 1
更新于2月 15, 2019
Fuse RPC op to improve the performance of distributed training 0 of 2 tasks completed

#13635 · opened 9月 27, 2018 by saxon_zh performance tuning
- 已关闭
- 2
更新于5月 22, 2020
the duration time of async recv op is imblance

#13566 · opened 9月 25, 2018 by saxon_zh performance tuning
- 已关闭
- 1
更新于5月 22, 2020
Refine parallel_do_grad

#9074 · opened 3月 14, 2018 by saxon_zh performance tuning
- 已关闭
- 0
更新于3月 15, 2018
The number of operations can be reduced to improve the performance.

#9045 · opened 3月 14, 2018 by saxon_zh performance tuning
- 已关闭
- 0
更新于6月 15, 2018
[Speed] Upgrade pybind from 2.1 to 2.2

#8944 · opened 3月 10, 2018 by saxon_zh performance tuning
- 已关闭
- 1
更新于3月 14, 2018
Optimize elementwise sub,div,min,max gradient functor

#8821 · opened 3月 07, 2018 by saxon_zh performance tuning
- 已关闭
- 0
更新于3月 08, 2018
GPU under-utilized when waiting for CPU to launch kernel

#8818 · opened 3月 07, 2018 by saxon_zh performance tuning
- 已关闭
- 7
更新于8月 15, 2018
ParallelDo performance on VGG

#8719 · opened 3月 03, 2018 by saxon_zh performance tuning
- 已关闭
- 2
更新于4月 21, 2018
Benchmark on convnets with multiple graphics card

#7958 · opened 1月 30, 2018 by saxon_zh MultiDevices performance tuning
- 已关闭
- 2
更新于4月 21, 2018
conv_2d cuDNN operator slow

#7902 · opened 1月 27, 2018 by saxon_zh performance tuning
- 已关闭
- 3
更新于4月 21, 2018
`elementwise_add_grad` should be optimized

#7862 · opened 1月 25, 2018 by saxon_zh performance tuning
- 已关闭
- 0
更新于2月 23, 2018
The LoD is pretty slow since it uses malloc/free/memcpy

#7204 · opened 1月 04, 2018 by saxon_zh performance tuning
- 已关闭
- 0
更新于1月 31, 2018
Speed data reader for IMDB dataset.

#7001 · opened 12月 25, 2017 by saxon_zh performance tuning
- 已关闭
- 0
更新于12月 26, 2017
`concat_op` is extremely slow in GPU

#6988 · opened 12月 25, 2017 by saxon_zh Release 0.11.1 performance tuning
- 已关闭
- 1
更新于2月 11, 2018
Adam operator takes too much time in CPU and lanch multiple kernels in CUDA

#6985 · opened 12月 25, 2017 by saxon_zh Release 0.11.1 performance tuning
- 已关闭
- 0
更新于12月 25, 2017

1
2
3