在Python中将矩阵列表转换为仅一个2D矩阵

Question

I have a list of 3960 matrices, which is simply the SIFT descriptors of 3960 images. This is supposed to result in a list of matrices with an unknown number of lines (which of course will depend on the image) and 128 columns (from SIFT descriptors). I am trying to put this list in just one 2D matrix, which the number of lines is the sum of the number of lines of these matrices and 128 columns, however, I am not being able to do that. Here is my code:

sift_keypoints = []

#read images from a text file
with open(file_images) as f:
    images_names = f.readlines()
    images_names = [a.strip() for a in images_names]

    for line in images_names:
        print(line)
        #read image
        image = cv2.imread(line,1)
        #Convert to grayscale
        image =cv2.cvtColor(image,cv2.COLOR_BGR2GRAY)
        #SIFT extraction
        sift = cv2.xfeatures2d.SIFT_create()
        kp, descriptors = sift.detectAndCompute(image,None)
        #appending sift keypoints to a list
        sift_keypoints.append(descriptors)

        #here is what I've tried
        sift_keypoints = np.asmatrix(np.asarray(sift_keypoints))

The sift_keypoints shape is (1,3960) according to this code, which is, of course, not what I want. How to transform this list in a 2D numpy array?

EDIT one simple example that illustrates my problem is the one in the following code

#how to convert this list to a matrix with shape (412,128)?
import numpy as np
x=np.zeros((256,128))
y=np.zeros((156,128))
list=[]
list.append(x)
list.append(y)

Answer 1

Solution using np.row_stack

Suppose l is your list of Numpy arrays of the shape (n, 128) .

Suppose that m is the total number of lines: the object is to stack all the objects and create a matrix of the shape (m, 128) .

We can proceed as follows, using Numpy's row_stack :

result = np.row_stack(l)

Answer 2

Use np.concatenate :

>>> from pprint import pprint
>>> import numpy as np
>>> 
>>> a = [np.full((2, 3), i) for i in range(3)]
>>> pprint(a)
[array([[0, 0, 0],
       [0, 0, 0]]),
 array([[1, 1, 1],
       [1, 1, 1]]),
 array([[2, 2, 2],                                                                                                  
       [2, 2, 2]])]                                                                                                 
>>> np.concatenate(a, axis=0)                                                                                       
array([[0, 0, 0],                                                                                                   
       [0, 0, 0],                                                                                                   
       [1, 1, 1],                                                                                                   
       [1, 1, 1],                                                                                                   
       [2, 2, 2],                                                                                                   
       [2, 2, 2]])