Efficient way of extracting from an array of objects

Question

I have a an array of objects in python:

meshnodearray = ['MeshNode object', 'MeshNode object', 'MeshNode object', ...]

Where for example first 'MeshNode object' is:

({'coordinates': (15.08, 273.01, 322.61), 'instanceName': None, 'label': 1})

I need to create an array of coordinates like this:

NODEcoo = np.zeros((nnod,3),dtype='float64')
for i in meshnodearray:
    NODEcoo[i.label-1,0:] = np.array(i.coordinates)

For large arrays this is slow. Is there a more efficient way of doing this, maybe without the for loop?

If you have to do that for all objects in the array then I see no other way. — cobie
– cobie, Commented May 5, 2012 at 11:26

alexis · Accepted Answer · 2012-05-05 17:53:55Z

1

Try extracting the coordinates into a python list of coordinates and converting it into a numpy array in one go. If the label values are sequential from 1 to nnod, it's as simple as this:

coords = [ n['coordinates'] for n in meshnodearray ]
NODEcoo = np.array(coords)

It would be somewhat better to do this with a generator (which would let you avoid creating the intermediate array), but numpy can create only one-dimensional arrays from a generator, with numpy.fromiter().

edited May 5, 2012 at 17:53

answered May 5, 2012 at 14:56

alexis

50.4k18 gold badges107 silver badges172 bronze badges

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

1 Comment

spule Over a year ago

Thanks, this is a little faster. Could this be done maybe in parallel to further speed it up?

Collectives™ on Stack Overflow

Efficient way of extracting from an array of objects

1 Answer 1

1 Comment

Your Answer

Hot Network Questions

Collectives™ on Stack Overflow

1 Answer 1

1 Comment

Your Answer

Sign up or log in

Post as a guest

Related