Python Forum

Suppose I have the following code:

import numpy as np
a=np.zeros((2),dtype=np.object)
b=np.array([[11],[13]])
a[0]=b[0]
a[1]=b[1]
print(a)

This gives me the wanted result of "[array([11]) array([13])]". But if I have a big array, it's too tedious to add a line for each index. So I wanna do it in some way like this:

a[0:2]=b[0:2]

But that gives me the error "ValueError: could not broadcast input array from shape (2,1) into shape (2)". So how do I have to write that one line to accomplish the same as the code at the beginning?

Shouldn't it be: a = np.zeros((2,), dtype=np.object)

(2) is an int, (2,) is s tuple

(Mar-31-2020, 06:52 PM)deanhystad Wrote: [ -> ]Shouldn't it be: a = np.zeros((2,), dtype=np.object)

(2) is an int, (2,) is s tuple

No, it shouldn't.

The problem is exactly described in error message: you can't broadcast shape 2,1 into shape 2.

array([0, 0], dtype=object)   # a
array([[11],                  # b
       [13]])

Before any advice it would be good to know what is the objective i.e. what do you want to accomplish.

(Apr-01-2020, 04:35 AM)perfringo Wrote: [ -> ]The problem is exactly described in error message: you can't broadcast shape 2,1 into shape 2.

But it obviously doesn't describe a solution, and a ton of research hasn't led me any nearer on a solution.

(Apr-01-2020, 04:35 AM)perfringo Wrote: [ -> ]Before any advice it would be good to know what is the objective i.e. what do you want to accomplish.

The exact same result that I said I got in the first post, but without having to do it one by one. It's fine with an array of a size 2, but what if I had a size 10, 1000 or even higher? Then part of my code would look like this:

a[0]=b[0]
a[1]=b[1]
a[2]=b[2]
a[3]=b[3]
a[4]=b[4]
a[5]=b[5]
...
a[997]=b[997]
a[998]=b[998]
a[999]=b[999]

But there must be an easier way of doing this, using slicing, or list comprehension perhaps. And that's what I can't figure out how to do.