Filling an array with arrays in numpy?

Question

A rather simple problem, and yet I dont seem to find any solution to it:

I am trying to use the differentiation of two arrays within an array and insert it as a new result into a new array... I hope the code clarifies this intention, even though it is not working, as python is telling me that it cannot combine a sequence with array calculation... How can I go around this problem? Thanks in advance!

MeteorNum = 5
#Coordinate Setup:

XPos = np.random.randint(0, 600, (MeteorNum,1))
YPos = np.random.randint(0, 800, (MeteorNum,1))

XYPos = np.hstack((XPos,YPos))
XYPos = np.vstack((XYPos, XYPos[0]))
print XYPos

#Distances:
r = np.zeros((MeteorNum,MeteorNum))

for i in range(0, MeteorNum):
    for x in range(0, MeteorNum):
        r[x,i] = (XYPos[x,:] - XYPos[i,:])
r = r*10**5
print r

and here the error I get:

runfile('C:/Users/Marco DS/Dropbox/1_TUDelft/4Q/AE1205 Python/python codes      2015/bonus assigments/week_5 Gravity simu.py', wdir='C:/Users/Marco   DS/Dropbox/1_TUDelft/4Q/AE1205 Python/python codes 2015/bonus assigments')
[[249 660]
 [256 605]
 [119 423]
 [422 398]
 [480 504]
 [249 660]]
Traceback (most recent call last):

  File "<ipython-input-57-4cd9cc29ece0>", line 1, in <module>
    runfile('C:/Users/Marco DS/Dropbox/1_TUDelft/4Q/AE1205 Python/python     codes 2015/bonus assigments/week_5 Gravity simu.py', wdir='C:/Users/Marco     DS/Dropbox/1_TUDelft/4Q/AE1205 Python/python codes 2015/bonus assigments')

  File "C:\Users\Marco DS\Anaconda\lib\site-packages\spyderlib\widgets\externalshell\sitecustomize.py", line 682, in runfile
execfile(filename, namespace)

File "C:\Users\Marco DS\Anaconda\lib\site-    packages\spyderlib\widgets\externalshell\sitecustomize.py", line 71, in execfile
exec(compile(scripttext, filename, 'exec'), glob, loc)

  File "C:/Users/Marco DS/Dropbox/1_TUDelft/4Q/AE1205 Python/python codes 2015/bonus assigments/week_5 Gravity simu.py", line 37, in <module>
r[x,i] = (XYPos[x,:] - XYPos[i,:])

ValueError: setting an array element with a sequence.

Please show the entire error message (traceback), it should have a line number, the line it is stuck on, the function name, the module name, ... — Julien Spronck
– Julien Spronck, Commented May 20, 2015 at 13:43
What do you mean with differentiation? Just computing the difference of the left and the right column? — plonser
– plonser, Commented May 20, 2015 at 13:45
Is it intentional that you store the np.hstack and np.vstack in the same variable? — P.R.
– P.R., Commented May 20, 2015 at 13:45
P.R. the vstack is in order to copy and add the first element to the last element of the array. That way, when I use the for loops, I can make sure all values are included. — v2rwbtc6
– v2rwbtc6, Commented May 20, 2015 at 14:01

farhawa · Accepted Answer · 2015-05-20 14:08:02Z

2

try to add dtype= in the container array:

MeteorNum = 5
#Coordinate Setup:

XPos = nprand.randint(0, 600, (MeteorNum,1))
YPos = nprand.randint(0, 800, (MeteorNum,1))

XYPos = np.hstack((XPos,YPos))
XYPos = np.vstack((XYPos, XYPos[0]))
print XYPos

#Distances:
r = np.zeros((MeteorNum,MeteorNum),dtype=np.ndarray)

for i in range(0, MeteorNum):
    for x in range(0, MeteorNum):
        r[x,i] = (XYPos[x,:] - XYPos[i,:])
r = r*10**5
print r

EDIT

The dtype argument indicate the desired data-type for the array, e.g., numpy.int8. Default is numpy.float64.

edited May 20, 2015 at 14:08

answered May 20, 2015 at 13:45

farhawa

10.5k16 gold badges57 silver badges93 bronze badges

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

3 Comments

v2rwbtc6 Over a year ago

Yes! That worked for me! It now gives me a huge array, which is what I needed. However, all elements now also say "array" and then what is in it, but oh well. it does the job... What though does the dtype do?

farhawa Over a year ago

I edited my answer to define the dtype argument role

farhawa Over a year ago

@user3604362 you can accept the answer if it was helpful for you

Lee · Accepted Answer · 2015-05-20 14:08:14Z

1

To remove errors, I think you need an extra dimension, of size 2, in your array r:

r = np.zeros((MeteorNum,MeteorNum,2))

As XYPos[x,:] - XYPos[i,:] in your for loop has a shape of (2,)

This returns a single array of shape (5, 5, 2).

edited May 20, 2015 at 14:08

answered May 20, 2015 at 13:52

Lee

31.4k31 gold badges124 silver badges187 bronze badges

3 Comments

v2rwbtc6 Over a year ago

hmmm, no... this does work, however I end up with several arrays, whereas I only want one big array.

Lee Over a year ago

@user3604362 Really? r should be one array of shape (5, 5, 2). Use r.shape to check.

hpaulj Over a year ago

user, are you confusing a 3d array with several arrays? Do you understand the problem of trying to put 2 values into one slot of r?

Collectives™ on Stack Overflow

Filling an array with arrays in numpy?

2 Answers 2

3 Comments

3 Comments

Your Answer

Hot Network Questions

Collectives™ on Stack Overflow

2 Answers 2

3 Comments

3 Comments

Your Answer

Sign up or log in

Post as a guest

Related