Reshaping tensor 2d to 1d

tensor([[17,  0],
        [93,  0],
        [0,  0],
        [21,  0],
        [19,  0])

I want to remove 0 from this tensor, which is a two-dimensional array, and make it a one-dimensional array.

How can I make this tensor with the tensor below?

tensor([[17],
        [93],
        [0],
        [21],
        [19])

When using the code below, there is a problem that the existing zero disappears.
How should I fix this?

x = x.flatten()
x = x[x!=0]
x = np.reshape(x, ( -1, x.shape[0] ))

array([[17, 93, 21, 19]])

>Solution :

You can use slicing to index the 0 column

import torch

t = torch.tensor(
      [[17,  0],
       [93,  0],
       [0,  0],
       [21,  0],
       [19,  0]]
    )

print(t[:,0])

Output

tensor([17, 93,  0, 21, 19])

And if you want to keep it a 2D array then you can use numpy.reshape

import torch
import numpy as np

t = torch.tensor(
      [[17,  0],
       [93,  0],
       [0,  0],
       [21,  0],
       [19,  0]]
    )

print(np.reshape(t[:,0], (-1, 1)))

Output

array([[17],
       [93],
       [ 0],
       [21],
       [19]], dtype=int32)