Skip to content

Commit ebefdbe

Browse files
authored
Merge pull request #10791 from shanyi15/patch-1
add title for cpu_profiling.md
2 parents 85b6bb5 + 60f4b73 commit ebefdbe

File tree

2 files changed

+6
-2
lines changed

2 files changed

+6
-2
lines changed

doc/fluid/howto/optimization/cpu_profiling_cn.md

Lines changed: 3 additions & 1 deletion
Original file line numberDiff line numberDiff line change
@@ -1,3 +1,5 @@
1+
# CPU性能调优
2+
13
此教程会介绍如何使用Python的cProfile包、Python库yep、Google perftools来进行性能分析 (profiling) 与调优(performance tuning)。
24

35
Profling 指发现性能瓶颈。系统中的瓶颈可能和程序员开发过程中想象的瓶颈相去甚远。Tuning 指消除瓶颈。性能优化的过程通常是不断重复地 profiling 和 tuning。
@@ -8,7 +10,7 @@ PaddlePaddle 用户一般通过调用 Python API 编写深度学习程序。大
810
* Python 与 C++ 混合代码的性能分析
911

1012

11-
# Python代码的性能分析
13+
## Python代码的性能分析
1214

1315
### 生成性能分析文件
1416

doc/fluid/howto/optimization/cpu_profiling_en.md

Lines changed: 3 additions & 1 deletion
Original file line numberDiff line numberDiff line change
@@ -1,3 +1,5 @@
1+
# Tune CPU performance
2+
13
This tutorial introduces techniques we use to profile and tune the
24
CPU performance of PaddlePaddle. We will use Python packages
35
`cProfile` and `yep`, and Google's `perftools`.
@@ -14,7 +16,7 @@ the profiling and tuning of
1416
1. the Python code and
1517
1. the mixture of Python and C++ code.
1618

17-
# Profiling the Python Code
19+
## Profiling the Python Code
1820

1921
### Generate the Performance Profiling File
2022

0 commit comments

Comments
 (0)