Numpy Concatenate to empty array

Question

I am trying to concatenate a set of numpy arrays loaded from disk. All arrays have varying number of columns.

This is my code

import numpy as np

FILE_LIST = ["matrix a", "matrix b"]

result=np.array([[0,0],[0,0]]) # I need to avoid this zero matrix
for fileName in FILE_LIST:
    matrix = matrix= np.genfromtxt(fileName, delimiter=" ")    
    result = np.concatenate((result, matrix),axis=1) 

print result

Here I have initialised result to an array with zeroes as I cannot concatenate to an empty array. I need to avoid this zero array appended at the beginning of the result. How to achieve this?

plonser · Accepted Answer · 2015-04-02 11:00:49Z

2

I would recommend to first load all the data in an array and then apply numpys hstack in order to horizontally stack the arrays

result = np.hstack([np.genfromtxt(fileName,delimiter=" ") for fileName in FILE_LIST])

edited Apr 2, 2015 at 11:00

answered Apr 2, 2015 at 10:15

plonser

3,3632 gold badges21 silver badges23 bronze badges

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

Comments

hpaulj · Accepted Answer · 2015-04-02 16:43:48Z

0

It's not obvious why you need to avoid this. But you could do:

result=None
for fileName in FILE_LIST:
    matrix= np.genfromtxt(fileName, delimiter=" ")
    if result is None:
        result = matrix
    else:
        result = np.concatenate((result, matrix),axis=1)

Generally we try to avoid repeated concatenation (or append) to arrays, prefering instead to append to lists. But in this case genfromtxt is a big enough operation that it doesn't matter much how you combine the arrays.

With a list, the loop would be:

result=[]
for fileName in FILE_LIST:
    result.append(np.genfromtxt(fileName, delimiter=" "))    
result = np.concatenate(result,axis=1)

A list comprehension is essentially the same thing.

answered Apr 2, 2015 at 16:43

hpaulj

233k14 gold badges260 silver badges392 bronze badges

Collectives™ on Stack Overflow

Numpy Concatenate to empty array

2 Answers 2

Comments

Comments

Your Answer

Hot Network Questions

Collectives™ on Stack Overflow

2 Answers 2

Comments

Comments

Your Answer

Sign up or log in

Post as a guest

Related