creating numpy arrays in a for loop

Question

I would like to be able to create numpy arrays based on a number that could change.
For example, say I have 50 text files containing a 2x2 set of numbers

I would like to load those 50 files as numpy arrays and use them later in the code. A sample of code may look like:

import load numpy as np
num = 50 #this could change based on different conditions
for i in arange(num):
 data%d % i = np.loadtxt("datafromafile%d.txt" % i)

Is this something like this possible? Thanks

• Oliver W.: Are the arrays in those files all of the same shape (2x2) or can they differ?

Answer 1

You can store them in a list:

data = list()
for i in arange(num):
 data.append(np.loadtxt("datafromafile%d.txt" % i))

Then you can access each array with:

>>> data[0] + data[1] # sum of the first and second numpy array

• getaglow: elyase, thank you, I have used your fix many times since you submitted. Sorry it took so long to let you know how helpful you were.

Answer 2

As a oneliner it would be:

NUM = 50
data = [np.loadtxt("datafromafile%d.txt" % value) for value in np.arange(NUM)]

or

FILES = ['file1', 'file2', 'file3']
data = {key: np.loadtxt(key) for key in FILES}

as a dict with the filename as key.