Moving the first element to end in Numpy array

user_10577507 0
Viewed 1843

I have a NumPy array, which I want to move the first element to the end. I know how to do it by converting to the list and do the insert operation but I want to avoid this and do it in a better and optimized way.

Suppose this is my array and I want to move 0 to the end:

x = [[ 0 388.13 19.346 412.38 39.594]
    [ 0 388.15 8.5168 416 38.972] 
    [ 0 223.14 156.21 317.91 193.46]]

I know I can do it in this way by converting to the list:

X = X.tolist()
for ele in x:
    ele.insert(len(ele), ele.pop(0))

Is there any way to do it without converting it to a list (using numpy)? and also, the way I'm doing is not efficient, What is a possible way to make my code efficient in term of speed even by converting it to list?

Expected output:

x = [[388.13 19.346 412.38 39.594 0]
    [388.15 8.5168 416 38.972 0] 
    [223.14 156.21 317.91 193.46 0]]
python-3.x numpy python
2 Answers

Try np.roll :np.roll(x, -1, axis=1).tolist()

user_30448250

you could also try np.hstack:

np.hstack([x[:, 1:], x[:, :1]])

or even np.concatenate:

np.concatenate([x[:, 1:], x[:, :1]],  axis=1)

or even np.c_

np.c_[x[:, 1:], x[:, :1]]

all the above gives you:

array([[388.13  ,  19.346 , 412.38  ,  39.594 ,   0.    ],
       [388.15  ,   8.5168, 416.    ,  38.972 ,   0.    ],
       [223.14  , 156.21  , 317.91  , 193.46  ,   0.    ]])
user_83802720
Related