I want to fill an array with some number, the output should be like this:
[[x1, y1, z1],
[x1, y2, z2],
[x2, y1, z3],
[x2, y2, z4],
[x3, y1, z5],
[x3, y2, z6]]
I have the data x, y, and z (lists) with 3 values x, 2 values of y and 6 values of z inside the data.
I tried to create a new array and then fill it with for or while loop, but I can't get it fully filled.
How can I do it in numpy with python without filling it manually?
Using np.repeat and np.tile on x and y respectively to get the desired columns and then stacking them together with z into one array:
import numpy as np
x = [1, 2, 3]
y = [4, 5]
z = [6, 7, 8, 9, 10, 11]
a = np.array([np.repeat(x, 2), np.tile(y, 3), z]).T
result:
array([[ 1, 4, 6],
[ 1, 5, 7],
[ 2, 4, 8],
[ 2, 5, 9],
[ 3, 4, 10],
[ 3, 5, 11]])
For a pure Python solution, to work with any list size.
x = [11,12,13,14,15]
y = [21,23,24]
z = [31,32,33,34,35,36,37]
This verbose code:
tmp = []
for i in range(max(len(x), len(y), len(z))):
for yy in y:
tmp.append([x[i%len(x)], yy])
for i in range(len(tmp)):
tmp[i].append((z[i%len(z)]))
Returns
# [[11, 21, 31],
# [11, 23, 32],
# [11, 24, 33],
# [12, 21, 34],
# [12, 23, 35],
# [12, 24, 36],
# [13, 21, 37],
# [13, 23, 31],
# [13, 24, 32],
# [14, 21, 33],
# [14, 23, 34],
# [14, 24, 35],
# [15, 21, 36],
# [15, 23, 37],
# [15, 24, 31],
# [11, 21, 32],
# [11, 23, 33],
# [11, 24, 34],
# [12, 21, 35],
# [12, 23, 36],
# [12, 24, 37]]
You can achieve this using NumPy's meshgrid function to create a grid of all combinations of your x, y, and z values. Then, you can use numpy.column_stack to stack these arrays horizontally.
import numpy as np
# Your data
x = [1, 2, 3]
y = [1, 2]
z = [1, 2, 3, 4, 5, 6]
# Create a grid of all combinations
x_grid, y_grid, z_grid = np.meshgrid(x, y, z, indexing='ij')
# Reshape the grids to 1D arrays
x_flat = x_grid.flatten()
y_flat = y_grid.flatten()
z_flat = z_grid.flatten()
# Stack the arrays horizontally to create the final array
result_array = np.column_stack((x_flat, y_flat, z_flat))
print(result_array)
Suppose X = [x_1,x_2,...,x_n], Y = [y_1,y_2,...,x_m] and Z=[z_1,z_2,...,z_nm], you want to:
Pairing X and Y can be done using the product method from itertools, as in:
from itertools import product
X=[1,2]
Y=[3,4,5]
print(list(product(X,Y)))
This will return [(1, 3), (1, 4), (1, 5), (2, 3), (2, 4), (2, 5)]
The itertools library doesn't return a list, but an iterable, so we can compose these iterables.
Now we have XY = [(x1,y_1), (x1,y_2), ... (x_1,y_m), (x_2, y_1), (x_2, y_2), ... , (x_2, y_m), ..., (x_n,y_m)] with n*m elements, and we want to pair them, respectively to Z=[z_1,z_2, ..., z_nm], to do that we can use a builtin function called zip which will pair them.
from itertools import product
X=[1,2]
Y=[3,4,5]
Z="abcdef"
XY=product(X,Y)
print(list(zip(XY, Z)))
The last code will return [((1, 3), 'a'), ((1, 4), 'b'), ((1, 5), 'c'), ((2, 3), 'd'), ((2, 4), 'e'), ((2, 5), 'f')]. There is a slight deviation in the form: entries are on the form ((x,y),z) instead of [x,y,z], but we can fix that using a list comprehension:
from itertools import product
X=[1,2]
Y=[3,4,5]
Z="abcdef"
print([[x,y,z] for (x,y), z in zip(product(X,Y), Z)])
[[a1 b1] [a2 b2] ... [a5 b5]]. In my case, I want to pair each x with each y. Also, my data actually much larger than 2 x and 2 y