Select multiple elements from python array

Question

How can I select multiple elements from a python array? I know that this is possible in numpy array but I cannot use numpy array in this case. I want to select certain elements of an array similar to masked array usage but i get following error in python

 nonzero = numpyarray.nonzero()
 pythonarray[nonzero] = numpyarray[nonzero]
 *** TypeError: only integer arrays with one element can be converted to an index

numpyarray[nonzero] works fine but I cannot access pythonarray[nonzero]. Is there a way to do this in python array?

millimoose · Accepted Answer · 2012-09-05 02:03:02Z

1

Just use a loop:

for idx in nonzero:
    pythonarray[idx] = numpyarray[idx]

Numpy arrays probably support n-ary indices to maximize performance; Python lists don't go that far with optimisations so you have to use the simpler approach.

edited Sep 5, 2012 at 2:03

answered Sep 5, 2012 at 1:54

millimoose

40.1k11 gold badges89 silver badges138 bronze badges

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

11 Comments

Alok Over a year ago

Thanks. I guess this will slow down the performance a lot. I have to do this 1 million times. And oh btw, it should be idx in the assignment expression instead of nonzero

millimoose Over a year ago

@Alok If you need to do something a million times quickly, Python in general might not be a good choice. (If you can't keep this in C code using numpy.)

millimoose Over a year ago

@Alok: Are you sure that Python lists resize faster than numpy arrays? As far as I know Python lists are backed by C arrays, so they have to be reallocated and possibly copied as well. If you need to grow efficiently to a large size without a reasonable upper bound, I'd consider using some sort of C-based linked list implementation to get rid of all the reallocations.

millimoose Over a year ago

@Alok Also, if every line has exactly 30 elements, what you could do is have a Python / linked list of 30-element numpy arrays, one per line. (The IO cost should dominate the overhead of a linked list.) Once you read the whole file, you know how many elements there are. You could then either create one huge numpy array for the result and copy stuff over once from the list, or just use the list-of-arrays directly. (Which one is better depends on your use case.)

jfs Over a year ago

@Alok: you could ask a separate question about how to read millions lines appending 30 elements afterwards efficiently (provide a small working example even if it is too slow).

|

Collectives™ on Stack Overflow

Select multiple elements from python array

1 Answer 1

11 Comments

Your Answer

Linked

Hot Network Questions

Collectives™ on Stack Overflow

1 Answer 1

11 Comments

Your Answer

Sign up or log in

Post as a guest

Linked

Related