Commit 5a7f5ce

committed

correct some mistakes

1 parent efc49c2 commit 5a7f5ceCopy full SHA for 5a7f5ce

File tree

-3

lines changed

-3

lines changed

Lines changed: 4 additions & 3 deletions

Original file line number	Diff line number	Diff line change
`@@ -1,8 +1,9 @@`
`1`	`1`	`<!-- TOC -->`
`2`	`2`
`3`		`-- [NumPy 是什么?](#numpy-%E6%98%AF%E4%BB%80%E4%B9%88%EF%BC%9F)`
	`3`	`+- [NumPy 是什么?](#numpy-%E6%98%AF%E4%BB%80%E4%B9%88)`
`4`	`4`
`5`	`5`	`<!-- /TOC -->`
	`6`	`+`
`6`	`7`	`# NumPy 是什么?`
`7`	`8`
`8`	`9`	`> 原文:[What is NumPy?](http://docs.scipy.org/doc/numpy-dev/user/whatisnumpy.html)`
`@@ -26,15 +27,15 @@ for i in range(len(a)):`
`26`	`27`
`27`	`28`	`这就产生了正确的答案,但如果 a 和 b 都含有数以百万计的数字,我们将为Python的低效循环付出代价。我们可以这样以C语言编写代码来完成同样的任务(为清楚起见我们忽略变量声明和初始化、内存分配等):`
`28`	`29`
`29`		-```python
	`30`	+```C
`30`	`31`	`for (i = 0; i < rows; i++): {`
`31`	`32`	`c[i] = a[i]*b[i];`
`32`	`33`	`}`
`33`	`34`	```
`34`	`35`
`35`	`36`	`这节省了所有涉及解释Python代码和操作Python对象的开销,但没有了使用Python编码的优势。此外,编码所需的工作量随数据维数的增加而增加。例如对于一个二维数组,C代码(像上面一样简写)会扩展为:`
`36`	`37`
`37`		-```python
	`38`	+```C
`38`	`39`	`for (i = 0; i < rows; i++): {`
`39`	`40`	`for (j = 0; j < columns; j++): {`
`40`	`41`	`c[i][j] = a[i][j]*b[i][j];`

Comments

(0)