python 矩陣

矩陣是二維數(shù)組的特殊情況，其中每個數(shù)據(jù)元素具有嚴格相同的大小。所以每個矩陣也是一個二維數(shù)組，但反之亦然。矩陣是許多數(shù)學和科學計算中非常重要的數(shù)據(jù)結構。正如我們在前一章中已經(jīng)討論過的兩個雙維數(shù)組結構，我們將在本章中專注于矩陣特有的數(shù)據(jù)結構操作。

我們也使用numpy包進行矩陣數(shù)據(jù)操作。

矩陣示例

考慮在早上，中午，晚上和深夜測量記錄溫度1周的情況。它可以使用數(shù)組以及numpy中可用的重塑方法呈現(xiàn)為7x5矩陣。

from numpy import *
a = array([['mon',18,20,22,17],['tue',11,18,21,18],
           ['wed',15,21,20,19],['thu',11,20,22,21],
           ['fri',18,17,23,22],['sat',12,22,20,18],
           ['sun',13,15,19,16]])

m = reshape(a,(7,5))
print(m)

上述數(shù)據(jù)可以表示為如下的二維數(shù)組。

[['mon' '18' '20' '22' '17']
 ['tue' '11' '18' '21' '18']
 ['wed' '15' '21' '20' '19']
 ['thu' '11' '20' '22' '21']
 ['fri' '18' '17' '23' '22']
 ['sat' '12' '22' '20' '18']
 ['sun' '13' '15' '19' '16']]

訪問矩陣中的值

矩陣中的數(shù)據(jù)元素可以通過使用索引來訪問。訪問方法與在二維數(shù)組中訪問數(shù)據(jù)的方式相同。

from numpy import *
m = array([['mon',18,20,22,17],['tue',11,18,21,18],
           ['wed',15,21,20,19],['thu',11,20,22,21],
           ['fri',18,17,23,22],['sat',12,22,20,18],
           ['sun',13,15,19,16]])

# print data for wednesday
print(m[2])

# print data for friday evening
print(m[4][3])

當上面的代碼被執(zhí)行時，它會產(chǎn)生以下結果 -

['wed', 15, 21, 20, 19]
23

添加一行

from numpy import *
m = array([['mon',18,20,22,17],['tue',11,18,21,18],
           ['wed',15,21,20,19],['thu',11,20,22,21],
           ['fri',18,17,23,22],['sat',12,22,20,18],
           ['sun',13,15,19,16]])

m_r = append(m,[['avg',12,15,13,11]],0)

print(m_r)

當上面的代碼被執(zhí)行時，它會產(chǎn)生以下結果 -

[['mon' '18' '20' '22' '17']
 ['tue' '11' '18' '21' '18']
 ['wed' '15' '21' '20' '19']
 ['thu' '11' '20' '22' '21']
 ['fri' '18' '17' '23' '22']
 ['sat' '12' '22' '20' '18']
 ['sun' '13' '15' '19' '16']
 ['avg' '12' '15' '13' '11']]

添加一列

我們可以使用insert（）方法將列添加到矩陣。這里我們不得不提及我們想要添加列的索引以及包含添加的列的新值的數(shù)組。在下面的例子中，我們在開頭的第五個位置添加一個新列。

from numpy import *
m = array([['mon',18,20,22,17],['tue',11,18,21,18],
           ['wed',15,21,20,19],['thu',11,20,22,21],
           ['fri',18,17,23,22],['sat',12,22,20,18],
           ['sun',13,15,19,16]])

m_c = insert(m,[5],[[1],[2],[3],[4],[5],[6],[7]],1)

print(m_c)

當上面的代碼被執(zhí)行時，它會產(chǎn)生以下結果 -

[['mon' '18' '20' '22' '17' '1']
 ['tue' '11' '18' '21' '18' '2']
 ['wed' '15' '21' '20' '19' '3']
 ['thu' '11' '20' '22' '21' '4']
 ['fri' '18' '17' '23' '22' '5']
 ['sat' '12' '22' '20' '18' '6']
 ['sun' '13' '15' '19' '16' '7']]

從矩陣中刪除一行

我們可以使用delete（）方法從矩陣中刪除一行。我們必須指定行的索引以及行的值為0，列的值為1的軸值。

from numpy import *
m = array([['mon',18,20,22,17],['tue',11,18,21,18],
           ['wed',15,21,20,19],['thu',11,20,22,21],
           ['fri',18,17,23,22],['sat',12,22,20,18],
           ['sun',13,15,19,16]])

m = delete(m,[2],0)

print(m)

當上面的代碼被執(zhí)行時，它會產(chǎn)生以下結果 -

[['mon' '18' '20' '22' '17']
 ['tue' '11' '18' '21' '18']
 ['thu' '11' '20' '22' '21']
 ['fri' '18' '17' '23' '22']
 ['sat' '12' '22' '20' '18']
 ['sun' '13' '15' '19' '16']]

從matrix中刪除一列

我們可以使用delete（）方法從矩陣中刪除一列。我們必須指定列的索引以及一行為0，一列為1的軸值。

from numpy import *
m = array([['mon',18,20,22,17],['tue',11,18,21,18],
           ['wed',15,21,20,19],['thu',11,20,22,21],
           ['fri',18,17,23,22],['sat',12,22,20,18],
           ['sun',13,15,19,16]])

m = delete(m,s_[2],1)

print(m)

當上面的代碼被執(zhí)行時，它會產(chǎn)生以下結果 -

[['mon' '18' '22' '17']
 ['tue' '11' '21' '18']
 ['wed' '15' '20' '19']
 ['thu' '11' '22' '21']
 ['fri' '18' '23' '22']
 ['sat' '12' '20' '18']
 ['sun' '13' '19' '16']]

更新matrix中的一行

要更新矩陣行中的值，我們只需在行的索引處重新分配值。在下面的例子中，星期幾數(shù)據(jù)的所有值都標記為零。該行的索引是3。

from numpy import *
m = array([['mon',18,20,22,17],['tue',11,18,21,18],
           ['wed',15,21,20,19],['thu',11,20,22,21],
           ['fri',18,17,23,22],['sat',12,22,20,18],
           ['sun',13,15,19,16]])

m[3] = ['thu',0,0,0,0]

print(m)

當上面的代碼被執(zhí)行時，它會產(chǎn)生以下結果 -

[['mon' '18' '20' '22' '17']
 ['tue' '11' '18' '21' '18']
 ['wed' '15' '21' '20' '19']
 ['thu' '0' '0' '0' '0']
 ['fri' '18' '17' '23' '22']
 ['sat' '12' '22' '20' '18']
 ['sun' '13' '15' '19' '16']]