Parse data from external file for matplotlib bar chart

Question

I would like to read data from a text file which has two columns. This is a follow-up to a previously asked question Generating a matlibplot bar chart from two columns of data. While my question is similar to How to plot data from multiple two column text files with legends in Matplotlib?, I am unable to get my defined variables to work with a bar chart. The goal is to read data in from a file which contains two columns.

I first tried using genfromtxt which is commented out in the current code, but receive AttributeError: 'numpy.ndarray' object has no attribute 'split'. This is likely due to the data being read in an array type format:

>>> import numpy as np
>>> data = np.genfromtxt('input-file', delimiter = ' ')
>>> print(data)
[[ 72.   1.]
 [  9.   2.]
 [ 10.  36.]
 [ 74.   6.]
 [  0.  77.]
 [  5.   6.]
 [  6.  23.]
 [ 72.   1.]
 [  9.   2.]
 [ 10.  36.]
 [ 82.   1.]
 [ 74.   6.]
 [  0.  97.]
 [  5.   6.]
 [  6.  23.]
 [ 72.   1.]
 [  9.   2.]
 [ 10.  36.]
 [ 82.   1.]]

Next, I trid read data in directly but receive AttributeError: 'file' object has no attribute 'split'.

import numpy as np
import matplotlib.pyplot as plt

data = open('input-file', 'r')

#data = np.genfromtxt('input-file', delimiter = ' ')

counts = []
values = []

for line in data.split("\n"):
    x, y = line.split()
    values.append(int(x))
    counts.append(int(y))

#bar_width = 0.25
#opacity = 0.4

plt.bar(values, counts, color='g')
plt.ylabel('Frequency')
plt.xlabel('Length')
plt.title('Title')

plt.show()

Example contents from input-file.

72 1
9 2
10 36
74 6
0 77
5 6
6 23
72 1
9 2
10 36
82 1
74 6
0 97
5 6
6 23
72 1
9 2
10 36
82 1

Answer 1

file object doesn't have split function. You can read lines directly without split("\n")

for line in data:

Answer 2

What prevents you from directly plotting the columns of the numpy array returned from np.genfromtext?

data = np.genfromtxt('tmp.dat', delimiter = ' ')

plt.bar(data[:,0], data[:,1], color='g')
plt.ylabel('Frequency')
plt.xlabel('Length')
plt.title('Title')

plt.show()