Assigning numpy array elements through predefined array

Question

I have arrays newx and newy with size nx*ns and ny*ns respectively where nx!=ny.
I want to be able to set the elements defined by newx and newy in array f by:

f = np.zeros([nx,ny,ns])
for s in range(ns):
    f[newx[:,s],newy[:,s],s] = s

Unfortunately this gives an error:

ValueError: shape mismatch: objects cannot be broadcast to a single shape

I understand the error but for the life of me can't figure out the correct syntax. plz help out.

Edit: provided sample code:

import numpy as np

newx = np.array([[0,1],
                 [1,2],
                 [2,3],
                 [3,0]])
newy = np.array([[0,1],
                 [1,2],
                 [2,0]])

f = np.zeros([4,3,2])
for s in range(2):
    f[newx[:,s],newy[:,s],s] = s

newx and newy don't make any sense. If you want to set to s the elements at some coordinates, you should have that number of x,y,s triplets rather than a different number of x,s and y,s pairs. — Lorenzo Gatti
– Lorenzo Gatti, Commented Apr 8, 2014 at 15:11

George · Accepted Answer · 2014-04-08 14:50:09Z

1

newx and newy must have the same shape , so you must reshape them.

f = np.zeros([4,3,2])

newX = newx.reshape(8,1)
newY = newy.reshape(6,)

for s in range(2):
    f[newX , newY ,s] = s


print f

answered Apr 8, 2014 at 14:50

George

5,75117 gold badges92 silver badges176 bronze badges

Sign up to request clarification or add additional context in comments.

2 Comments

nluigi Over a year ago

Ok got it, your tip that the newx and newy should be the same size was the solution. thx

Daniel Over a year ago

Or better yet: np.tile(np.arange(2),(newx.shape[0],newy.shape[0],1)) will give the same result.

Collectives™ on Stack Overflow

Assigning numpy array elements through predefined array

1 Answer 1

2 Comments

Your Answer

Hot Network Questions

Collectives™ on Stack Overflow

1 Answer 1

2 Comments

Your Answer

Sign up or log in

Post as a guest

Related