Create a column vector with zeros in the front in python-CodePudding

I just switched from MATLAB to python and I want help in translating a piece of MATLAB code to python.

I have a 1000 x 4 matrix in which the values in the m'th column represents a signal.

For simplicity I am changing the dimensions to 5 x 4.

Let the values in the m'th column be [1,2,3,4,5].

I want to delay this by x samples that is if x=2 the delayed version will be [0,0,1,2,3].

Its essentially adding x zeros to the front and removing x values from the back.

So the output is a 5x1 column vector.

The MATLAB code is:

[zeros(2,1),signal(1:(end-2),m)];

The python code I wrote is:

[np.zeros((2,1)),signal[(length_of_signal-1)-2:,m]]

This doesn't seem to work and neither am I able to figure out what's wrong with it. Please do help.

CodePudding user response：

IIUC, try with vstack:

np.random.seed(100)

x = 2
m = 3
a = np.random.randint(1, 5, (5,4))

>>> np.vstack((np.zeros((x, 1)), a[:,3][:len(a)-x][np.newaxis].T))

array([[0.],
       [0.],
       [4.],
       [3.],
       [2.]])

Note that you need the np.newaxis because the transpose of a 1D array is still a 1D array (unlike in MATLAB, where the default array is 2D).