Numpy duplicate data

numpy Medium main use tile Functions and repeat Function to realize data duplication

tile function

tile The function is defined as follows ：

def tile(A, reps)

The function mainly has two parameters , Parameters A Specifies a duplicate array , Parameters reps The number of times the order is repeated .
Be careful ： If reps The length of is d, Then the dimension of the last repeated array is max(d, A.ndim), Take away d And A The maximum value of the dimension . If A.ndim < d, By adding a new axis A Turn into d dimension . For example, for shape by (3,) adopt 2 Dimensional repetition shape Turn into (1,3), adopt 3 Dimensional repetition shape Turn into (1,1,3).

Example ：

repeat function

The function is defined as follows ：

repeat(a, repeats, axis=None):

repeat The main functions are 3 Parameters , Parameters a Specify the array elements that need to be repeated , Parameters repeats Specify the number of repetitions , Parameters axis Specifies which axis to repeat along .

• axis=None, It will be flatten Current matrix , In fact, it becomes a line vector
• axis=0, Along y Axis copy , Actually increased the number of lines
• axis=1, Along x Axis copy , Actually increase the number of columns

>>> np.repeat(x, 3, axis=0)
array([[1, 2],
       [1, 2],
       [1, 2],
       [3, 4],
       [3, 4],
       [3, 4]])

summary ：
The main difference between these two functions is ：tile The function is to repeat the operation on the array ,repeat The function is to repeat the operation on each element in the array .