numpy教程：逻辑函数Logic functions

电玩女神 2022-08-04 14:58 128阅读 0赞

http://[blog.csdn.net/pipisorry/article/details/48208433][blog.csdn.net_pipisorry_article_details_48208433]

## 真值测试Truth value testing ##

<table> 
 <tbody> 
  <tr> 
   <td><a href="http://docs.scipy.org/doc/numpy/reference/generated/numpy.all.html#numpy.all" title="numpy.all" rel="nofollow"><span>all</span></a>(a[,&nbsp;axis,&nbsp;out,&nbsp;keepdims])</td> 
   <td>Test whether all array elements along a given axis evaluate to True.</td> 
  </tr> 
  <tr> 
   <td><a href="http://docs.scipy.org/doc/numpy/reference/generated/numpy.any.html#numpy.any" title="numpy.any" rel="nofollow"><span>any</span></a>(a[,&nbsp;axis,&nbsp;out,&nbsp;keepdims])</td> 
   <td>Test whether any array element along a given axis evaluates to True.</td> 
  </tr> 
 </tbody> 
</table>

只要数组中有一个值为True，则any()返回True；而只有数组的全部元素都为True，all()才返回True。

也可以直接当成array数组的方法使用。

### 判断numpy数组是否为空 ###

if a.size:  
 print('array is not empty')

如果通过python列表，把一个列表作为一个布尔值会产生True如果有项目，False如果它是空的。  
lst = \[\]  
if lst:  
 print "array has items"  
if not lst:  
 print "array is empty"

\[[Python的-如何检查数组不为空？][Python_-]\]

### 判断numpy数组中是否有True    ###

array.any()

[皮皮blog][blog]

## 数组内容Array contents ##

<table> 
 <tbody> 
  <tr> 
   <td><a href="http://docs.scipy.org/doc/numpy/reference/generated/numpy.isfinite.html#numpy.isfinite" title="numpy.isfinite" rel="nofollow"><span>isfinite</span></a>(x[,&nbsp;out])</td> 
   <td>Test element-wise for finiteness (not infinity or not Not a Number).</td> 
  </tr> 
  <tr> 
   <td><a href="http://docs.scipy.org/doc/numpy/reference/generated/numpy.isinf.html#numpy.isinf" title="numpy.isinf" rel="nofollow"><span>isinf</span></a>(x[,&nbsp;out])</td> 
   <td>Test element-wise for positive or negative infinity.</td> 
  </tr> 
  <tr> 
   <td><a href="http://docs.scipy.org/doc/numpy/reference/generated/numpy.isnan.html#numpy.isnan" title="numpy.isnan" rel="nofollow"><span>isnan</span></a>(x[,&nbsp;out])</td> 
   <td>Test element-wise for NaN and return result as a boolean array.</td> 
  </tr> 
  <tr> 
   <td><a href="http://docs.scipy.org/doc/numpy/reference/generated/numpy.isneginf.html#numpy.isneginf" title="numpy.isneginf" rel="nofollow"><span>isneginf</span></a>(x[,&nbsp;y])</td> 
   <td>Test element-wise for negative infinity, return result as bool array.</td> 
  </tr> 
  <tr> 
   <td><a href="http://docs.scipy.org/doc/numpy/reference/generated/numpy.isposinf.html#numpy.isposinf" title="numpy.isposinf" rel="nofollow"><span>isposinf</span></a>(x[,&nbsp;y])</td> 
   <td>Test element-wise for positive infinity, return result as bool array.</td> 
  </tr> 
 </tbody> 
</table>

### numpy.isnan ###

numpy判断一个元素是否为np.NaN，判断某元素是否是nan

numpy.isnan(element)

Note: 不能使用array\[0\] == np.NaN，总是返回False!

### numpy数组元素替换numpy.nan\_to\_num(x)    ###

判断某元素是否是nan,inf,neginf,如果是，nan换为0，inf换为一个非常大的数，neginf换为非常小的数

numpy.nan\_to\_num(x)  
Replace nan with zero and inf with finite numbers.  
Returns an array or scalar replacing Not a Number (NaN) with zero, (positive) infinity with a very large number and negative infinity with a very small (or negative) number.

## 数组类型测试Array type testing ##

<table> 
 <tbody> 
  <tr> 
   <td><a href="http://docs.scipy.org/doc/numpy/reference/generated/numpy.iscomplex.html#numpy.iscomplex" title="numpy.iscomplex" rel="nofollow"><span>iscomplex</span></a>(x)</td> 
   <td>Returns a bool array, where True if input element is complex.</td> 
  </tr> 
  <tr> 
   <td><a href="http://docs.scipy.org/doc/numpy/reference/generated/numpy.iscomplexobj.html#numpy.iscomplexobj" title="numpy.iscomplexobj" rel="nofollow"><span>iscomplexobj</span></a>(x)</td> 
   <td>Check for a complex type or an array of complex numbers.</td> 
  </tr> 
  <tr> 
   <td><a href="http://docs.scipy.org/doc/numpy/reference/generated/numpy.isfortran.html#numpy.isfortran" title="numpy.isfortran" rel="nofollow"><span>isfortran</span></a>(a)</td> 
   <td>Returns True if the array is Fortran contiguous but <em>not</em> C contiguous.</td> 
  </tr> 
  <tr> 
   <td><a href="http://docs.scipy.org/doc/numpy/reference/generated/numpy.isreal.html#numpy.isreal" title="numpy.isreal" rel="nofollow"><span>isreal</span></a>(x)</td> 
   <td>Returns a bool array, where True if input element is real.</td> 
  </tr> 
  <tr> 
   <td><a href="http://docs.scipy.org/doc/numpy/reference/generated/numpy.isrealobj.html#numpy.isrealobj" title="numpy.isrealobj" rel="nofollow"><span>isrealobj</span></a>(x)</td> 
   <td>Return True if x is a not complex type or an array of complex numbers.</td> 
  </tr> 
  <tr> 
   <td><a href="http://docs.scipy.org/doc/numpy/reference/generated/numpy.isscalar.html#numpy.isscalar" title="numpy.isscalar" rel="nofollow"><span>isscalar</span></a>(num)</td> 
   <td>Returns True if the type of <em>num</em> is a scalar type.</td> 
  </tr> 
 </tbody> 
</table>

## 逻辑操作Logical operations ##

<table> 
 <tbody> 
  <tr> 
   <td><a href="http://docs.scipy.org/doc/numpy/reference/generated/numpy.logical_and.html#numpy.logical_and" title="numpy.logical_and" rel="nofollow"><span>logical_and</span></a>(x1,&nbsp;x2[,&nbsp;out])</td> 
   <td>Compute the truth value of x1 AND x2 element-wise.</td> 
  </tr> 
  <tr> 
   <td><a href="http://docs.scipy.org/doc/numpy/reference/generated/numpy.logical_or.html#numpy.logical_or" title="numpy.logical_or" rel="nofollow"><span>logical_or</span></a>(x1,&nbsp;x2[,&nbsp;out])</td> 
   <td>Compute the truth value of x1 OR x2 element-wise.</td> 
  </tr> 
  <tr> 
   <td><a href="http://docs.scipy.org/doc/numpy/reference/generated/numpy.logical_not.html#numpy.logical_not" title="numpy.logical_not" rel="nofollow"><span>logical_not</span></a>(x[,&nbsp;out])</td> 
   <td>Compute the truth value of NOT x element-wise.</td> 
  </tr> 
  <tr> 
   <td><a href="http://docs.scipy.org/doc/numpy/reference/generated/numpy.logical_xor.html#numpy.logical_xor" title="numpy.logical_xor" rel="nofollow"><span>logical_xor</span></a>(x1,&nbsp;x2[,&nbsp;out])</td> 
   <td>Compute the truth value of x1 XOR x2, element-wise.</td> 
  </tr> 
 </tbody> 
</table>

### 两个0-1array相与操作    ###

判断两个0-1array有多少个相同的1， 两种方式

rate = np.count_nonzero(np.logical_and(fs_predict_array, ground_truth_array)) rate = np.count_nonzero(fs_predict_array * ground_truth_array)

不过fs\_predict\_array \* ground\_truth\_array返回的是0-1array，而np.logical\_and(fs\_predict\_array ,ground\_truth\_array)返回的是True-False array，但是都可以使用sum()得到1或者True的数目。

lz亲测下面的logical\_and操作运行速度更快，没有count\_nonzero会更快。

[皮皮blog][blog]

## 比较Comparison ##

<table> 
 <tbody> 
  <tr> 
   <td><a href="http://docs.scipy.org/doc/numpy/reference/generated/numpy.allclose.html#numpy.allclose" title="numpy.allclose" rel="nofollow"><span>allclose</span></a>(a,&nbsp;b[,&nbsp;rtol,&nbsp;atol,&nbsp;equal_nan])</td> 
   <td>Returns True if two arrays are element-wise equal within a tolerance.</td> 
  </tr> 
  <tr> 
   <td><a href="http://docs.scipy.org/doc/numpy/reference/generated/numpy.isclose.html#numpy.isclose" title="numpy.isclose" rel="nofollow"><span>isclose</span></a>(a,&nbsp;b[,&nbsp;rtol,&nbsp;atol,&nbsp;equal_nan])</td> 
   <td>Returns a boolean array where two arrays are element-wise equal within a tolerance.</td> 
  </tr> 
  <tr> 
   <td><a href="http://docs.scipy.org/doc/numpy/reference/generated/numpy.array_equal.html#numpy.array_equal" title="numpy.array_equal" rel="nofollow"><span>array_equal</span></a>(a1,&nbsp;a2)</td> 
   <td>True if two arrays have the same shape and elements, False otherwise.</td> 
  </tr> 
  <tr> 
   <td><a href="http://docs.scipy.org/doc/numpy/reference/generated/numpy.array_equiv.html#numpy.array_equiv" title="numpy.array_equiv" rel="nofollow"><span>array_equiv</span></a>(a1,&nbsp;a2)</td> 
   <td>Returns True if input arrays are shape consistent and all elements equal.</td> 
  </tr> 
 </tbody> 
</table>

<table>     
 <tbody> 
  <tr> 
   <td><a href="http://docs.scipy.org/doc/numpy/reference/generated/numpy.greater.html#numpy.greater" title="numpy.greater" rel="nofollow"><span>greater</span></a>(x1,&nbsp;x2[,&nbsp;out])</td> 
   <td>Return the truth value of (x1 &gt; x2) element-wise.</td> 
  </tr> 
  <tr> 
   <td><a href="http://docs.scipy.org/doc/numpy/reference/generated/numpy.greater_equal.html#numpy.greater_equal" title="numpy.greater_equal" rel="nofollow"><span>greater_equal</span></a>(x1,&nbsp;x2[,&nbsp;out])</td> 
   <td>Return the truth value of (x1 &gt;= x2) element-wise.</td> 
  </tr> 
  <tr> 
   <td><a href="http://docs.scipy.org/doc/numpy/reference/generated/numpy.less.html#numpy.less" title="numpy.less" rel="nofollow"><span>less</span></a>(x1,&nbsp;x2[,&nbsp;out])</td> 
   <td>Return the truth value of (x1 &lt; x2) element-wise.</td> 
  </tr> 
  <tr> 
   <td><a href="http://docs.scipy.org/doc/numpy/reference/generated/numpy.less_equal.html#numpy.less_equal" title="numpy.less_equal" rel="nofollow"><span>less_equal</span></a>(x1,&nbsp;x2[,&nbsp;out])</td> 
   <td>Return the truth value of (x1 =&lt; x2) element-wise.</td> 
  </tr> 
  <tr> 
   <td><a href="http://docs.scipy.org/doc/numpy/reference/generated/numpy.equal.html#numpy.equal" title="numpy.equal" rel="nofollow"><span>equal</span></a>(x1,&nbsp;x2[,&nbsp;out])</td> 
   <td>Return (x1 == x2) element-wise.</td> 
  </tr> 
  <tr> 
   <td><a href="http://docs.scipy.org/doc/numpy/reference/generated/numpy.not_equal.html#numpy.not_equal" title="numpy.not_equal" rel="nofollow"><span>not_equal</span></a>(x1,&nbsp;x2[,&nbsp;out])</td> 
   <td>Return (x1 != x2) element-wise.</td> 
  </tr> 
 </tbody> 
</table>

**allclose**

如果两个数组在tolerance误差范围内相等，则返回True。

from: [http://blog.csdn.net/pipisorry/article/details/48208433][blog.csdn.net_pipisorry_article_details_48208433]

ref: [Logic functions][]

[blog.csdn.net_pipisorry_article_details_48208433]: http://blog.csdn.net/pipisorry/article/details/48208433
[Python_-]: http://codego.net/261286/
[blog]: http://blog.csdn.net/pipisorry
[Logic functions]: http://docs.scipy.org/doc/numpy/reference/routines.logic.html