How to loop through a file of strings and load the string into a numpy ndarray

This will create a numpy vector of strings with a line per cell:

import numpy as np

with open("file.ext") as f:
    a = np.array(f.readlines())

This could be modified for, say, a file of floating-point space-separated values:

import numpy as np

with open("file.ext") as f:
    a = np.array([map(float, line.split()) for line in f])

Let's break down the argument to array() to clarify what's going on here.

• [line for line in f] would be equivalent to f.readlines() - it creates a list of strings, one per line in f.
• [line.split() for line in f] makes a 2D list of strings. Each line in f gets split at the spaces into a list of strings.
• Numpy's designed to deal with numeric values, though, not strings. So we need to turn each sublist of strings into a list of floats. map applies the same function to every element of a list (float in this case), so map(float, line.split()) will turn a list of strings into a list of floats.
• So [map(float, line.split()) for line in f] creates a list of lists of floats - one list per line, where the floats have been split up at the spaces. This then gets sent to array() which knows how to deal with a list of lists.

Also look into the genfromtxt and loadtxt family of Numpy functions.