Ndarray_2darray

import numpy as np

lista = [[10,20,30],[40,50,60]]
a = np.array(lista)
print(a) 

# [[10 20 30]
   [40 50 60]]

.
.

⌨️ flatten, ravel

• n차원 -> 1차원
• 차이점 :
  flatten은 copy본 사용 -> 원본이 손상 X
  ravel은 참조값으로 줌 -> 원본이 손상 O

a1 = a.flatten()
print(a1)    # [10 20 30 40 50 60]

a1[2] = 100
print(a1)    # [10 20 100 40 50 60]
print(a)  
# [[10 20 30]
   [40 50 60]]

a2 = a.ravel()   # [10 20 30 40 50 60]
print(a2)   

a2[2] = 100
print(a2)        # [10 20 100 40 50 60]
print(a) 
# [[ 10  20 100]
   [ 40  50  60]]

.
.

⌨️ concatenate

a1 = np.arange(10,20).reshape(2,5)
a2 = np.arange(110,120).reshape(2,5)

print(a1)  
# [[10 11 12 13 14]
   [15 16 17 18 19]]

print(a2)
# [[110 111 112 113 114]
   [115 116 117 118 119]]

print(np.concatenate((a1,a2),axis=0))
# [[ 10  11  12  13  14]
   [ 15  16  17  18  19]
   [110 111 112 113 114]
   [115 116 117 118 119]]

print(np.concatenate((a1,a2),axis=1))
# [[ 10  11  12  13  14 110 111 112 113 114]
   [ 15  16  17  18  19 115 116 117 118 119]]

• hstack / vstack`

  hstack : axis = 1
  vstack : axis = 0

print(np.hstack((a1,a2)))
# [[10  11  12  13  14 110 111 112 113 114]
   [15  16  17  18  19 115 116 117 118 119]]
   
print(np.vstack((a1,a2)))
[[ 10  11  12  13  14]
 [ 15  16  17  18  19]
 [110 111 112 113 114]
 [115 116 117 118 119]]

• hsplit / vsplit

hsplit

b = np.concatenate((a1,a1),axis=1)
print(b)
# [[ 10  11  12  13  14 110 111 112 113 114]
   [ 15  16  17  18  19 115 116 117 118 119]]
   
print(np.hsplit(b,5)) 
# [array([[10, 11],
          [15, 16]]), array([[12, 13],
          [17, 18]]), array([[ 14, 110],
          [ 19, 115]]), array([[111, 112],
          [116, 117]]), array([[113, 114],
          [118, 119]])]

vsplit

b2 = b.T
print(b2)
# [[ 10  15]
   [ 11  16]
   [ 12  17]
   [ 13  18]
   [ 14  19]
   [110 115]
   [111 116]
   [112 117]
   [113 118]
   [114 119]]
   
print(np.vsplit(b2,5))
# [array([[10, 15],
         [11, 16]]), array([[12, 17],
         [13, 18]]), array([[ 14,  19],
         [110, 115]]), array([[111, 116],
         [112, 117]]), array([[113, 118],
         [114, 119]])]

⌨️ np.sum

: np.sum(array, axis, keepdims)

• axis
  

a1 = np.arange(30).reshape(5,6)
print(a1)

# result
# [[ 0  1  2  3  4  5]
   [ 6  7  8  9 10 11]
   [12 13 14 15 16 17]
   [18 19 20 21 22 23]
   [24 25 26 27 28 29]]

» axis = 1 을 따라 sum

np.sum(a1, axis = 1)
# result
# array([ 15,  51,  87, 123, 159])

» axis = 0 을 따라 sum

np.sum(a1, axis=0)
# result
# array([60, 65, 70, 75, 80, 85])